CompanyKG is a heterogeneous graph consisting of 1,169,931 nodes and 50,815,503 undirected edges, with each node representing a real-world company and each edge signifying a relationship between the connected pair of companies.

Edges: We model 15 different inter-company relations as undirected edges, each of which corresponds to a unique edge type. These edge types capture various forms of similarity between connected company pairs. Associated with each edge of a certain type, we calculate a real-numbered weight as an approximation of the similarity level of that type. It is important to note that the constructed edges do not represent an exhaustive list of all possible edges due to incomplete information. Consequently, this leads to a sparse and occasionally skewed distribution of edges for individual relation/edge types. Such characteristics pose additional challenges for downstream learning tasks. Please refer to our paper for a detailed definition of edge types and weight calculations.

Nodes: The graph includes all companies connected by edges defined previously. Each node represents a company and is associated with a descriptive text, such as "Klarna is a fintech company that provides support for direct and post-purchase payments ...". To comply with privacy and confidentiality requirements, we encoded the text into numerical embeddings using four different pre-trained text embedding models: mSBERT (multilingual Sentence BERT), ADA2, SimCSE (fine-tuned on the raw company descriptions) and PAUSE.

Evaluation Tasks. The primary goal of CompanyKG is to develop algorithms and models for quantifying the similarity between pairs of companies. In order to evaluate the effectiveness of these methods, we have carefully curated three evaluation tasks:

Similarity Prediction (SP). To assess the accuracy of pairwise company similarity, we constructed the SP evaluation set comprising 3,219 pairs of companies that are labeled either as positive (similar, denoted by "1") or negative (dissimilar, denoted by "0"). Of these pairs, 1,522 are positive and 1,697 are negative.

Competitor Retrieval (CR). Each sample contains one target company and one of its direct competitors. It contains 76 distinct target companies, each of which has 5.3 competitors annotated in average. For a given target company A with N direct competitors in this CR evaluation set, we expect a competent method to retrieve all N competitors when searching for similar companies to A.

Similarity Ranking (SR) is designed to assess the ability of any method to rank candidate companies (numbered 0 and 1) based on their similarity to a query company. Paid human annotators, with backgrounds in engineering, science, and investment, were tasked with determining which candidate company is more similar to the query company. It resulted in an evaluation set comprising 1,856 rigorously labeled ranking questions. We retained 20% (368 samples) of this set as a validation set for model development.

Edge Prediction (EP) evaluates a model's ability to predict future or missing relationships between companies, providing forward-looking insights for investment professionals. The EP dataset, derived (and sampled) from new edges collected between April 6, 2023, and May 25, 2024, includes 40,000 samples, with edges not present in the pre-existing CompanyKG (a snapshot up until April 5, 2023).

Background and Motivation

In the investment industry, it is often essential to identify similar companies for a variety of purposes, such as market/competitor mapping and Mergers & Acquisitions (M&A). Identifying comparable companies is a critical task, as it can inform investment decisions, help identify potential synergies, and reveal areas for growth and improvement. The accurate quantification of inter-company similarity, also referred to as company similarity quantification, is the cornerstone to successfully executing such tasks. However, company similarity quantification is often a challenging and time-consuming process, given the vast amount of data available on each company, and the complex and diversified relationships among them.

While there is no universally agreed definition of company similarity, researchers and practitioners in PE industry have adopted various criteria to measure similarity, typically reflecting the companies' operations and relationships. These criteria can embody one or more dimensions such as industry sectors, employee profiles, keywords/tags, customers' review, financial performance, co-appearance in news, and so on. Investment professionals usually begin with a limited number of companies of interest (a.k.a. seed companies) and require an algorithmic approach to expand their search to a larger list of companies for potential investment.

In recent years, transformer-based Language Models (LMs) have become the preferred method for encoding textual company descriptions into vector-space embeddings. Then companies that are similar to the seed companies can be searched in the embedding space using distance metrics like cosine similarity. The rapid advancements in Large LMs (LLMs), such as GPT-3/4 and LLaMA, have significantly enhanced the performance of general-purpose conversational models. These models, such as ChatGPT, can be employed to answer questions related to similar company discovery and quantification in a Q&A format.

However, graph is still the most natural choice for representing and learning diverse company relations due to its ability to model complex relationships between a large number of entities. By representing companies as nodes and their relationships as edges, we can form a Knowledge Graph (KG). Utilizing this KG allows us to efficiently capture and analyze the network structure of the business landscape. Moreover, KG-based approaches allow us to leverage powerful tools from network science, graph theory, and graph-based machine learning, such as Graph Neural Networks (GNNs), to extract insights and patterns to facilitate similar company analysis. While there are various company datasets (mostly commercial/proprietary and non-relational) and graph datasets available (mostly for single link/node/graph-level predictions), there is a scarcity of datasets and benchmarks that combine both to create a large-scale KG dataset expressing rich pairwise company relations.

Source Code and Tutorial:https://github.com/llcresearch/CompanyKG2

Paper: to be published

This dataset compiles the top 2500 datasets from Kaggle, encompassing a diverse range of topics and contributors. It provides insights into dataset creation, usability, popularity, and more, offering valuable information for researchers, analysts, and data enthusiasts.

Research Analysis: Researchers can utilize this dataset to analyze trends in dataset creation, popularity, and usability scores across various categories.

Contributor Insights: Kaggle contributors can explore the dataset to gain insights into factors influencing the success and engagement of their datasets, aiding in optimizing future submissions.

Machine Learning Training: Data scientists and machine learning enthusiasts can use this dataset to train models for predicting dataset popularity or usability based on features such as creator, category, and file types.

Market Analysis: Analysts can leverage the dataset to conduct market analysis, identifying emerging trends and popular topics within the data science community on Kaggle.

Educational Purposes: Educators and students can use this dataset to teach and learn about data analysis, visualization, and interpretation within the context of real-world datasets and community-driven platforms like Kaggle.

Column Definitions:

Dataset Name: Name of the dataset. Created By: Creator(s) of the dataset. Last Updated in number of days: Time elapsed since last update. Usability Score: Score indicating the ease of use. Number of File: Quantity of files included. Type of file: Format of files (e.g., CSV, JSON). Size: Size of the dataset. Total Votes: Number of votes received. Category: Categorization of the dataset's subject matter.

Clustering is a widely used technique with a long and rich history in a variety of areas. However, most existing algorithms do not scale well to large datasets, or are missing theoretical guarantees of convergence. This article introduces a provably robust clustering algorithm based on loss minimization that performs well on Gaussian mixture models with outliers. It provides theoretical guarantees that the algorithm obtains high accuracy with high probability under certain assumptions. Moreover, it can also be used as an initialization strategy for k-means clustering. Experiments on real-world large-scale datasets demonstrate the effectiveness of the algorithm when clustering a large number of clusters, and a k-means algorithm initialized by the algorithm outperforms many of the classic clustering methods in both speed and accuracy, while scaling well to large datasets such as ImageNet. Supplementary materials for this article are available online.

In a large network of computers or wireless sensors, each of the components (henceforth, peers) has some data about the global state of the system. Much of the system's functionality such as message routing, information retrieval and load sharing relies on modeling the global state. We refer to the outcome of the function (e.g., the load experienced by each peer) as the emph{model} of the system. Since the state of the system is constantly changing, it is necessary to keep the models up-to-date. Computing global data mining models e.g. decision trees, k-means clustering in large distributed systems may be very costly due to the scale of the system and due to communication cost, which may be high. The cost further increases in a dynamic scenario when the data changes rapidly. In this paper we describe a two step approach for dealing with these costs. First, we describe a highly efficient emph{local} algorithm which can be used to monitor a wide class of data mining models. Then, we use this algorithm as a feedback loop for the monitoring of complex functions of the data such as its k-means clustering. The theoretical claims are corroborated with a thorough experimental analysis.

This dataset contains metadata (title, abstract, date of publication, field, etc) for around 1 million academic articles. Each record contains additional information on the country of study and whether the article makes use of data. Machine learning tools were used to classify the country of study and data use.

Our data source of academic articles is the Semantic Scholar Open Research Corpus (S2ORC) (Lo et al. 2020). The corpus contains more than 130 million English language academic papers across multiple disciplines. The papers included in the Semantic Scholar corpus are gathered directly from publishers, from open archives such as arXiv or PubMed, and crawled from the internet.

We placed some restrictions on the articles to make them usable and relevant for our purposes. First, only articles with an abstract and parsed PDF or latex file are included in the analysis. The full text of the abstract is necessary to classify the country of study and whether the article uses data. The parsed PDF and latex file are important for extracting important information like the date of publication and field of study. This restriction eliminated a large number of articles in the original corpus. Around 30 million articles remain after keeping only articles with a parsable (i.e., suitable for digital processing) PDF, and around 26% of those 30 million are eliminated when removing articles without an abstract. Second, only articles from the year 2000 to 2020 were considered. This restriction eliminated an additional 9% of the remaining articles. Finally, articles from the following fields of study were excluded, as we aim to focus on fields that are likely to use data produced by countries’ national statistical system: Biology, Chemistry, Engineering, Physics, Materials Science, Environmental Science, Geology, History, Philosophy, Math, Computer Science, and Art. Fields that are included are: Economics, Political Science, Business, Sociology, Medicine, and Psychology. This third restriction eliminated around 34% of the remaining articles. From an initial corpus of 136 million articles, this resulted in a final corpus of around 10 million articles.

Due to the intensive computer resources required, a set of 1,037,748 articles were randomly selected from the 10 million articles in our restricted corpus as a convenience sample.

The empirical approach employed in this project utilizes text mining with Natural Language Processing (NLP). The goal of NLP is to extract structured information from raw, unstructured text. In this project, NLP is used to extract the country of study and whether the paper makes use of data. We will discuss each of these in turn.

To determine the country or countries of study in each academic article, two approaches are employed based on information found in the title, abstract, or topic fields. The first approach uses regular expression searches based on the presence of ISO3166 country names. A defined set of country names is compiled, and the presence of these names is checked in the relevant fields. This approach is transparent, widely used in social science research, and easily extended to other languages. However, there is a potential for exclusion errors if a country’s name is spelled non-standardly.

The second approach is based on Named Entity Recognition (NER), which uses machine learning to identify objects from text, utilizing the spaCy Python library. The Named Entity Recognition algorithm splits text into named entities, and NER is used in this project to identify countries of study in the academic articles. SpaCy supports multiple languages and has been trained on multiple spellings of countries, overcoming some of the limitations of the regular expression approach. If a country is identified by either the regular expression search or NER, it is linked to the article. Note that one article can be linked to more than one country.

The second task is to classify whether the paper uses data. A supervised machine learning approach is employed, where 3500 publications were first randomly selected and manually labeled by human raters using the Mechanical Turk service (Paszke et al. 2019).[1] To make sure the human raters had a similar and appropriate definition of data in mind, they were given the following instructions before seeing their first paper:

Each of these documents is an academic article. The goal of this study is to measure whether a specific academic article is using data and from which country the data came.

There are two classification tasks in this exercise:

1. identifying whether an academic article is using data from any country

2. Identifying from which country that data came.

For task 1, we are looking specifically at the use of data. Data is any information that has been collected, observed, generated or created to produce research findings. As an example, a study that reports findings or analysis using a survey data, uses data. Some clues to indicate that a study does use data includes whether a survey or census is described, a statistical model estimated, or a table or means or summary statistics is reported.

After an article is classified as using data, please note the type of data used. The options are population or business census, survey data, administrative data, geospatial data, private sector data, and other data. If no data is used, then mark "Not applicable". In cases where multiple data types are used, please click multiple options.[2]

For task 2, we are looking at the country or countries that are studied in the article. In some cases, no country may be applicable. For instance, if the research is theoretical and has no specific country application. In some cases, the research article may involve multiple countries. In these cases, select all countries that are discussed in the paper.

We expect between 10 and 35 percent of all articles to use data.

The median amount of time that a worker spent on an article, measured as the time between when the article was accepted to be classified by the worker and when the classification was submitted was 25.4 minutes. If human raters were exclusively used rather than machine learning tools, then the corpus of 1,037,748 articles examined in this study would take around 50 years of human work time to review at a cost of $3,113,244, which assumes a cost of $3 per article as was paid to MTurk workers.

A model is next trained on the 3,500 labelled articles. We use a distilled version of the BERT (bidirectional Encoder Representations for transformers) model to encode raw text into a numeric format suitable for predictions (Devlin et al. (2018)). BERT is pre-trained on a large corpus comprising the Toronto Book Corpus and Wikipedia. The distilled version (DistilBERT) is a compressed model that is 60% the size of BERT and retains 97% of the language understanding capabilities and is 60% faster (Sanh, Debut, Chaumond, Wolf 2019). We use PyTorch to produce a model to classify articles based on the labeled data. Of the 3,500 articles that were hand coded by the MTurk workers, 900 are fed to the machine learning model. 900 articles were selected because of computational limitations in training the NLP model. A classification of “uses data” was assigned if the model predicted an article used data with at least 90% confidence.

The performance of the models classifying articles to countries and as using data or not can be compared to the classification by the human raters. We consider the human raters as giving us the ground truth. This may underestimate the model performance if the workers at times got the allocation wrong in a way that would not apply to the model. For instance, a human rater could mistake the Republic of Korea for the Democratic People’s Republic of Korea. If both humans and the model perform the same kind of errors, then the performance reported here will be overestimated.

The model was able to predict whether an article made use of data with 87% accuracy evaluated on the set of articles held out of the model training. The correlation between the number of articles written about each country using data estimated under the two approaches is given in the figure below. The number of articles represents an aggregate total of

Overview

3DHD CityScenes is the most comprehensive, large-scale high-definition (HD) map dataset to date, annotated in the three spatial dimensions of globally referenced, high-density LiDAR point clouds collected in urban domains. Our HD map covers 127 km of road sections of the inner city of Hamburg, Germany including 467 km of individual lanes. In total, our map comprises 266,762 individual items.

Our corresponding paper (published at ITSC 2022) is available here. Further, we have applied 3DHD CityScenes to map deviation detection here.

Moreover, we release code to facilitate the application of our dataset and the reproducibility of our research. Specifically, our 3DHD_DevKit comprises:

Python tools to read, generate, and visualize the dataset,

3DHDNet deep learning pipeline (training, inference, evaluation) for map deviation detection and 3D object detection.

The DevKit is available here:

https://github.com/volkswagen/3DHD_devkit.

The dataset and DevKit have been created by Christopher Plachetka as project lead during his PhD period at Volkswagen Group, Germany.

When using our dataset, you are welcome to cite:

@INPROCEEDINGS{9921866, author={Plachetka, Christopher and Sertolli, Benjamin and Fricke, Jenny and Klingner, Marvin and Fingscheidt, Tim}, booktitle={2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC)}, title={3DHD CityScenes: High-Definition Maps in High-Density Point Clouds}, year={2022}, pages={627-634}}

Acknowledgements

We thank the following interns for their exceptional contributions to our work.

Benjamin Sertolli: Major contributions to our DevKit during his master thesis

Niels Maier: Measurement campaign for data collection and data preparation

The European large-scale project Hi-Drive (www.Hi-Drive.eu) supports the publication of 3DHD CityScenes and encourages the general publication of information and databases facilitating the development of automated driving technologies.

The Dataset

After downloading, the 3DHD_CityScenes folder provides five subdirectories, which are explained briefly in the following.

1. Dataset

This directory contains the training, validation, and test set definition (train.json, val.json, test.json) used in our publications. Respective files contain samples that define a geolocation and the orientation of the ego vehicle in global coordinates on the map.

During dataset generation (done by our DevKit), samples are used to take crops from the larger point cloud. Also, map elements in reach of a sample are collected. Both modalities can then be used, e.g., as input to a neural network such as our 3DHDNet.

To read any JSON-encoded data provided by 3DHD CityScenes in Python, you can use the following code snipped as an example.

import json

json_path = r"E:\3DHD_CityScenes\Dataset\train.json" with open(json_path) as jf: data = json.load(jf) print(data)

1. HD_Map

Map items are stored as lists of items in JSON format. In particular, we provide:

traffic signs,

traffic lights,

pole-like objects,

construction site locations,

construction site obstacles (point-like such as cones, and line-like such as fences),

line-shaped markings (solid, dashed, etc.),

polygon-shaped markings (arrows, stop lines, symbols, etc.),

lanes (ordinary and temporary),

relations between elements (only for construction sites, e.g., sign to lane association).

1. HD_Map_MetaData

Our high-density point cloud used as basis for annotating the HD map is split in 648 tiles. This directory contains the geolocation for each tile as polygon on the map. You can view the respective tile definition using QGIS. Alternatively, we also provide respective polygons as lists of UTM coordinates in JSON.

Files with the ending .dbf, .prj, .qpj, .shp, and .shx belong to the tile definition as “shape file” (commonly used in geodesy) that can be viewed using QGIS. The JSON file contains the same information provided in a different format used in our Python API.

1. HD_PointCloud_Tiles

The high-density point cloud tiles are provided in global UTM32N coordinates and are encoded in a proprietary binary format. The first 4 bytes (integer) encode the number of points contained in that file. Subsequently, all point cloud values are provided as arrays. First all x-values, then all y-values, and so on. Specifically, the arrays are encoded as follows.

x-coordinates: 4 byte integer

y-coordinates: 4 byte integer

z-coordinates: 4 byte integer

intensity of reflected beams: 2 byte unsigned integer

ground classification flag: 1 byte unsigned integer

After reading, respective values have to be unnormalized. As an example, you can use the following code snipped to read the point cloud data. For visualization, you can use the pptk package, for instance.

import numpy as np import pptk

file_path = r"E:\3DHD_CityScenes\HD_PointCloud_Tiles\HH_001.bin" pc_dict = {} key_list = ['x', 'y', 'z', 'intensity', 'is_ground'] type_list = ['

Traditional k-means and most k-means variants are still computationally expensive for large datasets, such as microarray data, which have large datasets with large dimension size d. In k-means clustering, we are given a set of n data points in d-dimensional space Rd and an integer k. The problem is to determine a set of k points in Rd, called centers, so as to minimize the mean squared distance from each data point to its nearest center. In this work, we develop a novel k-means algorithm, which is simple but more efficient than the traditional k-means and the recent enhanced k-means. Our new algorithm is based on the recently established relationship between principal component analysis and the k-means clustering. We provided the correctness proof for this algorithm. Results obtained from testing the algorithm on three biological data and six non-biological data (three of these data are real, while the other three are simulated) also indicate that our algorithm is empirically faster than other known k-means algorithms. We assessed the quality of our algorithm clusters against the clusters of a known structure using the Hubert-Arabie Adjusted Rand index (ARIHA). We found that when k is close to d, the quality is good (ARIHA>0.8) and when k is not close to d, the quality of our new k-means algorithm is excellent (ARIHA>0.9). In this paper, emphases are on the reduction of the time requirement of the k-means algorithm and its application to microarray data due to the desire to create a tool for clustering and malaria research. However, the new clustering algorithm can be used for other clustering needs as long as an appropriate measure of distance between the centroids and the members is used. This has been demonstrated in this work on six non-biological data.

Digitization of healthcare data along with algorithmic breakthroughts in AI will have a major impact on healthcare delivery in coming years. Its intresting to see application of AI to assist clinicians during patient treatment in a privacy preserving way. While scientific knowledge can help guide interventions, there remains a key need to quickly cut through the space of decision policies to find effective strategies to support patients during the care process.

Offline Reinforcement learning (also referred to as safe or batch reinforcement learning) is a promising sub-field of RL which provides us with a mechanism for solving real world sequential decision making problems where access to simulator is not available. Here we assume that learn a policy from fixed dataset of trajectories with further interaction with the environment(agent doesn't receive reward or punishment signal from the environment). It has shown that such an approach can leverage vast amount of existing logged data (in the form of previous interactions with the environment) and can outperform supervised learning approaches or heuristic based policies for solving real world - decision making problems. Offline RL algorithms when trained on sufficiently large and diverse offline datasets can produce close to optimal policies(ability to generalize beyond training data).

As Part of my PhD, research, I investigated the problem of developing a Clinical Decision Support System for Sepsis Management using Offline Deep Reinforcement Learning.

MIMIC-III ('Medical Information Mart for Intensive Care') is a large open-access anonymized single-center database which consists of comprehensive clinical data of 61,532 critical care admissions from 2001–2012 collected at a Boston teaching hospital. Dataset consists of 47 features (including demographics, vitals, and lab test results) on a cohort of sepsis patients who meet the sepsis-3 definition criteria.

we try to answer the following question:

Given a particular patient’s characteristics and physiological information at each time step as input, can our DeepRL approach, learn an optimal treatment policy that can prescribe the right intervention(e.g use of ventilator) to the patient each stage of the treatment process, in order to improve the final outcome(e.g patient mortality)?

we can use popular state-of-the-art algorithms such as Deep Q Learning(DQN), Double Deep Q Learning (DDQN), DDQN combined with BNC, Mixed Monte Carlo(MMC) and Persistent Advantage Learning (PAL). Using these methods we can train an RL policy to recommend optimum treatment path for a given patient.

Data acquisition, standard pre-processing and modelling details can be found here in Github repo: https://github.com/asjad99/MIMIC_RL_COACH

General

For more details and the most up-to-date information please consult our project page: https://kainmueller-lab.github.io/fisbe.

Summary

A new dataset for neuron instance segmentation in 3d multicolor light microscopy data of fruit fly brains

30 completely labeled (segmented) images

71 partly labeled images

altogether comprising ∼600 expert-labeled neuron instances (labeling a single neuron takes between 30-60 min on average, yet a difficult one can take up to 4 hours)

To the best of our knowledge, the first real-world benchmark dataset for instance segmentation of long thin filamentous objects

A set of metrics and a novel ranking score for respective meaningful method benchmarking

An evaluation of three baseline methods in terms of the above metrics and score

Abstract

Instance segmentation of neurons in volumetric light microscopy images of nervous systems enables groundbreaking research in neuroscience by facilitating joint functional and morphological analyses of neural circuits at cellular resolution. Yet said multi-neuron light microscopy data exhibits extremely challenging properties for the task of instance segmentation: Individual neurons have long-ranging, thin filamentous and widely branching morphologies, multiple neurons are tightly inter-weaved, and partial volume effects, uneven illumination and noise inherent to light microscopy severely impede local disentangling as well as long-range tracing of individual neurons. These properties reflect a current key challenge in machine learning research, namely to effectively capture long-range dependencies in the data. While respective methodological research is buzzing, to date methods are typically benchmarked on synthetic datasets. To address this gap, we release the FlyLight Instance Segmentation Benchmark (FISBe) dataset, the first publicly available multi-neuron light microscopy dataset with pixel-wise annotations. In addition, we define a set of instance segmentation metrics for benchmarking that we designed to be meaningful with regard to downstream analyses. Lastly, we provide three baselines to kick off a competition that we envision to both advance the field of machine learning regarding methodology for capturing long-range data dependencies, and facilitate scientific discovery in basic neuroscience.

Dataset documentation:

We provide a detailed documentation of our dataset, following the Datasheet for Datasets questionnaire:

FISBe Datasheet

Our dataset originates from the FlyLight project, where the authors released a large image collection of nervous systems of ~74,000 flies, available for download under CC BY 4.0 license.

Files

fisbe_v1.0_{completely,partly}.zip

contains the image and ground truth segmentation data; there is one zarr file per sample, see below for more information on how to access zarr files.

fisbe_v1.0_mips.zip

maximum intensity projections of all samples, for convenience.

sample_list_per_split.txt

a simple list of all samples and the subset they are in, for convenience.

view_data.py

a simple python script to visualize samples, see below for more information on how to use it.

dim_neurons_val_and_test_sets.json

a list of instance ids per sample that are considered to be of low intensity/dim; can be used for extended evaluation.

Readme.md

general information

How to work with the image files

Each sample consists of a single 3d MCFO image of neurons of the fruit fly.For each image, we provide a pixel-wise instance segmentation for all separable neurons.Each sample is stored as a separate zarr file (zarr is a file storage format for chunked, compressed, N-dimensional arrays based on an open-source specification.").The image data ("raw") and the segmentation ("gt_instances") are stored as two arrays within a single zarr file.The segmentation mask for each neuron is stored in a separate channel.The order of dimensions is CZYX.

We recommend to work in a virtual environment, e.g., by using conda:

conda create -y -n flylight-env -c conda-forge python=3.9conda activate flylight-env

How to open zarr files

Install the python zarr package:

pip install zarr

Opened a zarr file with:

import zarrraw = zarr.open(, mode='r', path="volumes/raw")seg = zarr.open(, mode='r', path="volumes/gt_instances")

optional:import numpy as npraw_np = np.array(raw)

Zarr arrays are read lazily on-demand.Many functions that expect numpy arrays also work with zarr arrays.Optionally, the arrays can also explicitly be converted to numpy arrays.

How to view zarr image files

We recommend to use napari to view the image data.

Install napari:

pip install "napari[all]"

Save the following Python script:

import zarr, sys, napari

raw = zarr.load(sys.argv[1], mode='r', path="volumes/raw")gts = zarr.load(sys.argv[1], mode='r', path="volumes/gt_instances")

viewer = napari.Viewer(ndisplay=3)for idx, gt in enumerate(gts): viewer.add_labels( gt, rendering='translucent', blending='additive', name=f'gt_{idx}')viewer.add_image(raw[0], colormap="red", name='raw_r', blending='additive')viewer.add_image(raw[1], colormap="green", name='raw_g', blending='additive')viewer.add_image(raw[2], colormap="blue", name='raw_b', blending='additive')napari.run()

Execute:

python view_data.py /R9F03-20181030_62_B5.zarr

Metrics

S: Average of avF1 and C

avF1: Average F1 Score

C: Average ground truth coverage

clDice_TP: Average true positives clDice

FS: Number of false splits

FM: Number of false merges

tp: Relative number of true positives

For more information on our selected metrics and formal definitions please see our paper.

Baseline

To showcase the FISBe dataset together with our selection of metrics, we provide evaluation results for three baseline methods, namely PatchPerPix (ppp), Flood Filling Networks (FFN) and a non-learnt application-specific color clustering from Duan et al..For detailed information on the methods and the quantitative results please see our paper.

License

The FlyLight Instance Segmentation Benchmark (FISBe) dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

Citation

If you use FISBe in your research, please use the following BibTeX entry:

@misc{mais2024fisbe, title = {FISBe: A real-world benchmark dataset for instance segmentation of long-range thin filamentous structures}, author = {Lisa Mais and Peter Hirsch and Claire Managan and Ramya Kandarpa and Josef Lorenz Rumberger and Annika Reinke and Lena Maier-Hein and Gudrun Ihrke and Dagmar Kainmueller}, year = 2024, eprint = {2404.00130}, archivePrefix ={arXiv}, primaryClass = {cs.CV} }

Acknowledgments

We thank Aljoscha Nern for providing unpublished MCFO images as well as Geoffrey W. Meissner and the entire FlyLight Project Team for valuablediscussions.P.H., L.M. and D.K. were supported by the HHMI Janelia Visiting Scientist Program.This work was co-funded by Helmholtz Imaging.

Changelog

There have been no changes to the dataset so far.All future change will be listed on the changelog page.

Contributing

If you would like to contribute, have encountered any issues or have any suggestions, please open an issue for the FISBe dataset in the accompanying github repository.

All contributions are welcome!

This constitute simulation of three large data sets in the order of; 10,000×50, 30,000×50 and 50,000×50 dimension. The range of K used is 10≤K≤40 for the four algorithms.

LScDC Word-Category RIG MatrixApril 2020 by Neslihan Suzen, PhD student at the University of Leicester (ns433@leicester.ac.uk / suzenneslihan@hotmail.com)Supervised by Prof Alexander Gorban and Dr Evgeny MirkesGetting StartedThis file describes the Word-Category RIG Matrix for theLeicester Scientific Corpus (LSC) [1], the procedure to build the matrix and introduces the Leicester Scientific Thesaurus (LScT) with the construction process. The Word-Category RIG Matrix is a 103,998 by 252 matrix, where rows correspond to words of Leicester Scientific Dictionary-Core (LScDC) [2] and columns correspond to 252 Web of Science (WoS) categories [3, 4, 5]. Each entry in the matrix corresponds to a pair (category,word). Its value for the pair shows the Relative Information Gain (RIG) on the belonging of a text from the LSC to the category from observing the word in this text. The CSV file of Word-Category RIG Matrix in the published archive is presented with two additional columns of the sum of RIGs in categories and the maximum of RIGs over categories (last two columns of the matrix). So, the file ‘Word-Category RIG Matrix.csv’ contains a total of 254 columns.This matrix is created to be used in future research on quantifying of meaning in scientific texts under the assumption that words have scientifically specific meanings in subject categories and the meaning can be estimated by information gains from word to categories. LScT (Leicester Scientific Thesaurus) is a scientific thesaurus of English. The thesaurus includes a list of 5,000 words from the LScDC. We consider ordering the words of LScDC by the sum of their RIGs in categories. That is, words are arranged in their informativeness in the scientific corpus LSC. Therefore, meaningfulness of words evaluated by words’ average informativeness in the categories. We have decided to include the most informative 5,000 words in the scientific thesaurus. Words as a Vector of Frequencies in WoS CategoriesEach word of the LScDC is represented as a vector of frequencies in WoS categories. Given the collection of the LSC texts, each entry of the vector consists of the number of texts containing the word in the corresponding category.It is noteworthy that texts in a corpus do not necessarily belong to a single category, as they are likely to correspond to multidisciplinary studies, specifically in a corpus of scientific texts. In other words, categories may not be exclusive. There are 252 WoS categories and a text can be assigned to at least 1 and at most 6 categories in the LSC. Using the binary calculation of frequencies, we introduce the presence of a word in a category. We create a vector of frequencies for each word, where dimensions are categories in the corpus.The collection of vectors, with all words and categories in the entire corpus, can be shown in a table, where each entry corresponds to a pair (word,category). This table is build for the LScDC with 252 WoS categories and presented in published archive with this file. The value of each entry in the table shows how many times a word of LScDC appears in a WoS category. The occurrence of a word in a category is determined by counting the number of the LSC texts containing the word in a category. Words as a Vector of Relative Information Gains Extracted for CategoriesIn this section, we introduce our approach to representation of a word as a vector of relative information gains for categories under the assumption that meaning of a word can be quantified by their information gained for categories.For each category, a function is defined on texts that takes the value 1, if the text belongs to the category, and 0 otherwise. For each word, a function is defined on texts that takes the value 1 if the word belongs to the text, and 0 otherwise. Consider LSC as a probabilistic sample space (the space of equally probable elementary outcomes). For the Boolean random variables, the joint probability distribution, the entropy and information gains are defined.The information gain about the category from the word is the amount of information on the belonging of a text from the LSC to the category from observing the word in the text [6]. We used the Relative Information Gain (RIG) providing a normalised measure of the Information Gain. This provides the ability of comparing information gains for different categories. The calculations of entropy, Information Gains and Relative Information Gains can be found in the README file in the archive published. Given a word, we created a vector where each component of the vector corresponds to a category. Therefore, each word is represented as a vector of relative information gains. It is obvious that the dimension of vector for each word is the number of categories. The set of vectors is used to form the Word-Category RIG Matrix, in which each column corresponds to a category, each row corresponds to a word and each component is the relative information gain from the word to the category. In Word-Category RIG Matrix, a row vector represents the corresponding word as a vector of RIGs in categories. We note that in the matrix, a column vector represents RIGs of all words in an individual category. If we choose an arbitrary category, words can be ordered by their RIGs from the most informative to the least informative for the category. As well as ordering words in each category, words can be ordered by two criteria: sum and maximum of RIGs in categories. The top n words in this list can be considered as the most informative words in the scientific texts. For a given word, the sum and maximum of RIGs are calculated from the Word-Category RIG Matrix.RIGs for each word of LScDC in 252 categories are calculated and vectors of words are formed. We then form the Word-Category RIG Matrix for the LSC. For each word, the sum (S) and maximum (M) of RIGs in categories are calculated and added at the end of the matrix (last two columns of the matrix). The Word-Category RIG Matrix for the LScDC with 252 categories, the sum of RIGs in categories and the maximum of RIGs over categories can be found in the database.Leicester Scientific Thesaurus (LScT)Leicester Scientific Thesaurus (LScT) is a list of 5,000 words form the LScDC [2]. Words of LScDC are sorted in descending order by the sum (S) of RIGs in categories and the top 5,000 words are selected to be included in the LScT. We consider these 5,000 words as the most meaningful words in the scientific corpus. In other words, meaningfulness of words evaluated by words’ average informativeness in the categories and the list of these words are considered as a ‘thesaurus’ for science. The LScT with value of sum can be found as CSV file with the published archive. Published archive contains following files:1) Word_Category_RIG_Matrix.csv: A 103,998 by 254 matrix where columns are 252 WoS categories, the sum (S) and the maximum (M) of RIGs in categories (last two columns of the matrix), and rows are words of LScDC. Each entry in the first 252 columns is RIG from the word to the category. Words are ordered as in the LScDC.2) Word_Category_Frequency_Matrix.csv: A 103,998 by 252 matrix where columns are 252 WoS categories and rows are words of LScDC. Each entry of the matrix is the number of texts containing the word in the corresponding category. Words are ordered as in the LScDC.3) LScT.csv: List of words of LScT with sum (S) values. 4) Text_No_in_Cat.csv: The number of texts in categories. 5) Categories_in_Documents.csv: List of WoS categories for each document of the LSC.6) README.txt: Description of Word-Category RIG Matrix, Word-Category Frequency Matrix and LScT and forming procedures.7) README.pdf (same as 6 in PDF format)References[1] Suzen, Neslihan (2019): LSC (Leicester Scientific Corpus). figshare. Dataset. https://doi.org/10.25392/leicester.data.9449639.v2[2] Suzen, Neslihan (2019): LScDC (Leicester Scientific Dictionary-Core). figshare. Dataset. https://doi.org/10.25392/leicester.data.9896579.v3[3] Web of Science. (15 July). Available: https://apps.webofknowledge.com/[4] WoS Subject Categories. Available: https://images.webofknowledge.com/WOKRS56B5/help/WOS/hp_subject_category_terms_tasca.html [5] Suzen, N., Mirkes, E. M., & Gorban, A. N. (2019). LScDC-new large scientific dictionary. arXiv preprint arXiv:1912.06858. [6] Shannon, C. E. (1948). A mathematical theory of communication. Bell system technical journal, 27(3), 379-423.

This file contains the data set used to develop a random forest model predict background specific conductivity for stream segments in the contiguous United States. This Excel readable file contains 56 columns of parameters evaluated during development. The data dictionary provides the definition of the abbreviations and the measurement units. Each row is a unique sample described as R** which indicates the NHD Hydrologic Unit (underscore), up to a 7-digit COMID, (underscore) sequential sample month. To develop models that make stream-specific predictions across the contiguous United States, we used StreamCat data set and process (Hill et al. 2016; https://github.com/USEPA/StreamCat). The StreamCat data set is based on a network of stream segments from NHD+ (McKay et al. 2012). These stream segments drain an average area of 3.1 km2 and thus define the spatial grain size of this data set. The data set consists of minimally disturbed sites representing the natural variation in environmental conditions that occur in the contiguous 48 United States. More than 2.4 million SC observations were obtained from STORET (USEPA 2016b), state natural resource agencies, the U.S. Geological Survey (USGS) National Water Information System (NWIS) system (USGS 2016), and data used in Olson and Hawkins (2012) (Table S1). Data include observations made between 1 January 2001 and 31 December 2015 thus coincident with Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data (https://modis.gsfc.nasa.gov/data/). Each observation was related to the nearest stream segment in the NHD+. Data were limited to one observation per stream segment per month. SC observations with ambiguous locations and repeat measurements along a stream segment in the same month were discarded. Using estimates of anthropogenic stress derived from the StreamCat database (Hill et al. 2016), segments were selected with minimal amounts of human activity (Stoddard et al. 2006) using criteria developed for each Level II Ecoregion (Omernik and Griffith 2014). Segments were considered as potentially minimally stressed where watersheds had 0 - 0.5% impervious surface, 0 – 5% urban, 0 – 10% agriculture, and population densities from 0.8 – 30 people/km2 (Table S3). Watersheds with observations with large residuals in initial models were identified and inspected for evidence of other human activities not represented in StreamCat (e.g., mining, logging, grazing, or oil/gas extraction). Observations were removed from disturbed watersheds, with a tidal influence or unusual geologic conditions such as hot springs. About 5% of SC observations in each National Rivers and Stream Assessment (NRSA) region were then randomly selected as independent validation data. The remaining observations became the large training data set for model calibration. This dataset is associated with the following publication: Olson, J., and S. Cormier. Modeling spatial and temporal variation in natural background specific conductivity. ENVIRONMENTAL SCIENCE & TECHNOLOGY. American Chemical Society, Washington, DC, USA, 53(8): 4316-4325, (2019).

ERA5 is the fifth generation ECMWF reanalysis for the global climate and weather for the past 8 decades. Data is available from 1940 onwards. ERA5 replaces the ERA-Interim reanalysis. Reanalysis combines model data with observations from across the world into a globally complete and consistent dataset using the laws of physics. This principle, called data assimilation, is based on the method used by numerical weather prediction centres, where every so many hours (12 hours at ECMWF) a previous forecast is combined with newly available observations in an optimal way to produce a new best estimate of the state of the atmosphere, called analysis, from which an updated, improved forecast is issued. Reanalysis works in the same way, but at reduced resolution to allow for the provision of a dataset spanning back several decades. Reanalysis does not have the constraint of issuing timely forecasts, so there is more time to collect observations, and when going further back in time, to allow for the ingestion of improved versions of the original observations, which all benefit the quality of the reanalysis product. ERA5 provides hourly estimates for a large number of atmospheric, ocean-wave and land-surface quantities. An uncertainty estimate is sampled by an underlying 10-member ensemble at three-hourly intervals. Ensemble mean and spread have been pre-computed for convenience. Such uncertainty estimates are closely related to the information content of the available observing system which has evolved considerably over time. They also indicate flow-dependent sensitive areas. To facilitate many climate applications, monthly-mean averages have been pre-calculated too, though monthly means are not available for the ensemble mean and spread. ERA5 is updated daily with a latency of about 5 days (monthly means are available around the 6th of each month). In case that serious flaws are detected in this early release (called ERA5T), this data could be different from the final release 2 to 3 months later. So far this has only been the case for the month September 2021, while it will also be the case for October, November and December 2021. For months prior to September 2021 the final release has always been equal to ERA5T, and the goal is to align the two again after December 2021. ERA5 is updated daily with a latency of about 5 days (monthly means are available around the 6th of each month). In case that serious flaws are detected in this early release (called ERA5T), this data could be different from the final release 2 to 3 months later. In case that this occurs users are notified. The data set presented here is a regridded subset of the full ERA5 data set on native resolution. It is online on spinning disk, which should ensure fast and easy access. It should satisfy the requirements for most common applications. An overview of all ERA5 datasets can be found in this article. Information on access to ERA5 data on native resolution is provided in these guidelines. Data has been regridded to a regular lat-lon grid of 0.25 degrees for the reanalysis and 0.5 degrees for the uncertainty estimate (0.5 and 1 degree respectively for ocean waves). There are four main sub sets: hourly and monthly products, both on pressure levels (upper air fields) and single levels (atmospheric, ocean-wave and land surface quantities). The present entry is "ERA5 monthly mean data on pressure levels from 1940 to present".

How best can we summarise sets of phylogenetic trees? Systematists have relied heavily on consensus methods, but if tree distributions can be partitioned into distinct subsets it may be helpful to provide separate summaries of these rather than relying entirely upon a single consensus tree. How sets of trees can most helpfully be partitioned and represented leads to many open questions, but one natural partitioning is provided by the islands of trees found during tree searches. Islands that are of dissimilar size have been shown to yield majority-rule consensus trees dominated by the largest sets. We illustrate this large island bias and approaches that mitigate its impact by revisiting a recent analysis of phylogenetic relationships of living and fossil amphibians. We introduce a revised definition of tree islands based on any tree-to-tree pairwise distance metric that usefully extends the notion to any set or multiset of trees, as might be produced by, for example, Bayesian or bootst...

Abstract

The documentation covers Enterprise Survey panel datasets that were collected in Slovenia in 2009, 2013 and 2019.

The Slovenia ES 2009 was conducted between 2008 and 2009. The Slovenia ES 2013 was conducted between March 2013 and September 2013. Finally, the Slovenia ES 2019 was conducted between December 2018 and November 2019. The objective of the Enterprise Survey is to gain an understanding of what firms experience in the private sector.

As part of its strategic goal of building a climate for investment, job creation, and sustainable growth, the World Bank has promoted improving the business environment as a key strategy for development, which has led to a systematic effort in collecting enterprise data across countries. The Enterprise Surveys (ES) are an ongoing World Bank project in collecting both objective data based on firms' experiences and enterprises' perception of the environment in which they operate.

Geographic coverage

National

Analysis unit

The primary sampling unit of the study is the establishment. An establishment is a physical location where business is carried out and where industrial operations take place or services are provided. A firm may be composed of one or more establishments. For example, a brewery may have several bottling plants and several establishments for distribution. For the purposes of this survey an establishment must take its own financial decisions and have its own financial statements separate from those of the firm. An establishment must also have its own management and control over its payroll.

Universe

As it is standard for the ES, the Slovenia ES was based on the following size stratification: small (5 to 19 employees), medium (20 to 99 employees), and large (100 or more employees).

Kind of data

Sample survey data [ssd]

Sampling procedure

The sample for Slovenia ES 2009, 2013, 2019 were selected using stratified random sampling, following the methodology explained in the Sampling Manual for Slovenia 2009 ES and for Slovenia 2013 ES, and in the Sampling Note for 2019 Slovenia ES.

Three levels of stratification were used in this country: industry, establishment size, and oblast (region). The original sample designs with specific information of the industries and regions chosen are included in the attached Excel file (Sampling Report.xls.) for Slovenia 2009 ES. For Slovenia 2013 and 2019 ES, specific information of the industries and regions chosen is described in the "The Slovenia 2013 Enterprise Surveys Data Set" and "The Slovenia 2019 Enterprise Surveys Data Set" reports respectively, Appendix E.

For the Slovenia 2009 ES, industry stratification was designed in the way that follows: the universe was stratified into manufacturing industries, services industries, and one residual (core) sector as defined in the sampling manual. Each industry had a target of 90 interviews. For the manufacturing industries sample sizes were inflated by about 17% to account for potential non-response cases when requesting sensitive financial data and also because of likely attrition in future surveys that would affect the construction of a panel. For the other industries (residuals) sample sizes were inflated by about 12% to account for under sampling in firms in service industries.

For Slovenia 2013 ES, industry stratification was designed in the way that follows: the universe was stratified into one manufacturing industry, and two service industries (retail, and other services).

Finally, for Slovenia 2019 ES, three levels of stratification were used in this country: industry, establishment size, and region. The original sample design with specific information of the industries and regions chosen is described in "The Slovenia 2019 Enterprise Surveys Data Set" report, Appendix C. Industry stratification was done as follows: Manufacturing – combining all the relevant activities (ISIC Rev. 4.0 codes 10-33), Retail (ISIC 47), and Other Services (ISIC 41-43, 45, 46, 49-53, 55, 56, 58, 61, 62, 79, 95).

For Slovenia 2009 and 2013 ES, size stratification was defined following the standardized definition for the rollout: small (5 to 19 employees), medium (20 to 99 employees), and large (more than 99 employees). For stratification purposes, the number of employees was defined on the basis of reported permanent full-time workers. This seems to be an appropriate definition of the labor force since seasonal/casual/part-time employment is not a common practice, except in the sectors of construction and agriculture.

For Slovenia 2009 ES, regional stratification was defined in 2 regions. These regions are Vzhodna Slovenija and Zahodna Slovenija. The Slovenia sample contains panel data. The wave 1 panel “Investment Climate Private Enterprise Survey implemented in Slovenia” consisted of 223 establishments interviewed in 2005. A total of 57 establishments have been re-interviewed in the 2008 Business Environment and Enterprise Performance Survey.

For Slovenia 2013 ES, regional stratification was defined in 2 regions (city and the surrounding business area) throughout Slovenia.

Finally, for Slovenia 2019 ES, regional stratification was done across two regions: Eastern Slovenia (NUTS code SI03) and Western Slovenia (SI04).

Mode of data collection

Computer Assisted Personal Interview [capi]

Research instrument

Questionnaires have common questions (core module) and respectfully additional manufacturing- and services-specific questions. The eligible manufacturing industries have been surveyed using the Manufacturing questionnaire (includes the core module, plus manufacturing specific questions). Retail firms have been interviewed using the Services questionnaire (includes the core module plus retail specific questions) and the residual eligible services have been covered using the Services questionnaire (includes the core module). Each variation of the questionnaire is identified by the index variable, a0.

Response rate

Survey non-response must be differentiated from item non-response. The former refers to refusals to participate in the survey altogether whereas the latter refers to the refusals to answer some specific questions. Enterprise Surveys suffer from both problems and different strategies were used to address these issues.

Item non-response was addressed by two strategies: a- For sensitive questions that may generate negative reactions from the respondent, such as corruption or tax evasion, enumerators were instructed to collect the refusal to respond as (-8). b- Establishments with incomplete information were re-contacted in order to complete this information, whenever necessary. However, there were clear cases of low response.

For 2009 and 2013 Slovenia ES, the survey non-response was addressed by maximizing efforts to contact establishments that were initially selected for interview. Up to 4 attempts were made to contact the establishment for interview at different times/days of the week before a replacement establishment (with similar strata characteristics) was suggested for interview. Survey non-response did occur but substitutions were made in order to potentially achieve strata-specific goals. Further research is needed on survey non-response in the Enterprise Surveys regarding potential introduction of bias.

For 2009, the number of contacted establishments per realized interview was 6.18. This number is the result of two factors: explicit refusals to participate in the survey, as reflected by the rate of rejection (which includes rejections of the screener and the main survey) and the quality of the sample frame, as represented by the presence of ineligible units. The relatively low ratio of contacted establishments per realized interview (6.18) suggests that the main source of error in estimates in the Slovenia may be selection bias and not frame inaccuracy.

For 2013, the number of realized interviews per contacted establishment was 25%. This number is the result of two factors: explicit refusals to participate in the survey, as reflected by the rate of rejection (which includes rejections of the screener and the main survey) and the quality of the sample frame, as represented by the presence of ineligible units. The number of rejections per contact was 44%.

Finally, for 2019, the number of interviews per contacted establishments was 9.7%. This number is the result of two factors: explicit refusals to participate in the survey, as reflected by the rate of rejection (which includes rejections of the screener and the main survey) and the quality of the sample frame, as represented by the presence of ineligible units. The share of rejections per contact was 75.2%.

Spending a lot of time engaging in activities related to substance useUsing a substance in greater quantities or for a longer time than intendedDeveloping tolerance (i.e., needing to use the substance more than before to get desired effects or noticing that the same amount of substance use had less effect than before)Making unsuccessful attempts to cut down on useContinuing substance use despite physical health or emotional problems associated with substance useReducing or eliminating participation in other activities because of substance useExperiencing withdrawal symptoms.Similarly, Abuse is also defined consistent with the DSM-IV definition as the following lifestyle symptoms due to the use of illicit drugs in the past 12 months:Experiencing problems at work, home, and schoolDoing something physically dangerousExperiencing Repeated trouble with the law

TACRED is a large-scale relation extraction dataset with 106,264 examples built over newswire and web text from the corpus used in the yearly TAC Knowledge Base Population (TAC KBP) challenges. Examples in TACRED cover 41 relation types as used in the TAC KBP challenges (e.g., per:schools_attended and org:members) or are labeled as no_relation if no defined relation is held. These examples are created by combining available human annotations from the TAC KBP challenges and crowdsourcing.

The Hydrology Feature Dataset contains photogrammetrically compiled water drainage features and structures including rivers, streams, drainage canals, locks, dams, lakes, ponds, reservoirs and mooring cells. Rivers, Lakes, Ponds, Reservoirs, Hidden Lakes, Reservoirs or Ponds: If greater than 25 feet and less than 30 feet wide, is captured as a double line stream. If greater than 30 feet wide it is captured as a river. Lakes are large standing bodies of water greater than 5 acres in size. Ponds are large standing bodies of water greater than 1 acre and less than 5 acres in size. Polygons are created from Stream edges and River Edges. The Ohio River, Monongahela River and Allegheny River are coded as Major Rivers. All other River and Stream polygons are coded as River. If a stream is less than 25 feet wide it is placed as a single line and coded as a Stream. Both sides of the stream are digitized and coded as a Stream for Streams whose width is greater than 25 feet. River edges are digitized and coded as River.

A Drainage Canal is a manmade or channelized hydrographic feature. Drainage Canals are differentiated from streams in that drainage canals have had the sides and/or bottom stabilized to prevent erosion for the predominant length of the feature. Streams may have had some stabilization done, but are primarily in a natural state. Lakes are large standing bodies of water greater than five acres in size. Ponds are large standing bodies of water greater than one acre in size and less than five acres in size. Reservoirs are manmade embankments of water. Included in this definition are both covered and uncovered water tanks. Reservoirs that are greater than one acre in size are digitized. Hidden Streams, Hidden Rivers and Hidden Drainage Canal or Culverts are those areas of drainage where the water flows through a manmade facility such as a culvert. Hydrology Annotation is not being updated but will be preserved. If a drainage feature has been removed, as apparent on the aerial photography, the associated drainage name annotation will be removed. A Mooring Cell is a structure to which tows can tie off while awaiting lockage. They are normally constructed of concrete and steel and are anchored to the river bottom by means of gravity or sheet piling.

Mooring Cells do not currently exist in the Allegheny County dataset but will be added. Locks are devices that are used to control flow or access to a hydrologic feature. The edges of the Lock are captured. Dams are devices that are used to hold or delay the natural flow of water. The edges of the Dam are shown.

This dataset is harvested on a weekly basis from Allegheny County’s GIS data portal. The full metadata record for this dataset can also be found on Allegheny County's GIS portal. You can access the metadata record and other resources on the GIS portal by clicking on the “Explore” button (and choosing the "Go to resource" option) to the right of the "ArcGIS Open Dataset" text below.

Category: Environment

Department: Geographic Information Systems Group; Department of Administrative Services

Data Notes: Coordinate System: Pennsylvania State Plane South Zone 3702; U.S. Survey Foot

Development Notes: Original Lakes and Drainage datasets combined to create this layer. Data was updated as a result of a flyover in the spring of 2004. A database field has been defined for all map features named Update Year". This database field will define which dataset provided each map feature. Map features from the current map will be set to "2004". The earlier dataset map features the earlier dataset map features used to supplement the area near the county boundary will be set to "1993". All new or modified map data will have the value for "Update Year" set to "2004".

Data Dictionary: https://docs.google.com/spreadsheets/d/16BWrRkoPtq2ANRkrbG7CrfQk2dUsWRiaS2Ee1mTn7l0/edit?usp=sharing

Objective: To develop a clinical informatics pipeline designed to capture large-scale structured EHR data for a national patient registry.

Materials and Methods: The EHR-R-REDCap pipeline is implemented using R-statistical software to remap and import structured EHR data into the REDCap-based multi-institutional Merkel Cell Carcinoma (MCC) Patient Registry using an adaptable data dictionary.

Results: Clinical laboratory data were extracted from EPIC Clarity across several participating institutions. Labs were transformed, remapped and imported into the MCC registry using the EHR labs abstraction (eLAB) pipeline. Forty-nine clinical tests encompassing 482,450 results were imported into the registry for 1,109 enrolled MCC patients. Data-quality assessment revealed highly accurate, valid labs. Univariate modeling was performed for labs at baseline on overall survival (N=176) using this clinical informatics pipeline.

Conclusion: We demonstrate feasibility of the facile eLAB workflow. EHR data is successfully transformed, and bulk-loaded/imported into a REDCap-based national registry to execute real-world data analysis and interoperability.

Methods eLAB Development and Source Code (R statistical software):

eLAB is written in R (version 4.0.3), and utilizes the following packages for processing: DescTools, REDCapR, reshape2, splitstackshape, readxl, survival, survminer, and tidyverse. Source code for eLAB can be downloaded directly (https://github.com/TheMillerLab/eLAB).

eLAB reformats EHR data abstracted for an identified population of patients (e.g. medical record numbers (MRN)/name list) under an Institutional Review Board (IRB)-approved protocol. The MCCPR does not host MRNs/names and eLAB converts these to MCCPR assigned record identification numbers (record_id) before import for de-identification.

Functions were written to remap EHR bulk lab data pulls/queries from several sources including Clarity/Crystal reports or institutional EDW including Research Patient Data Registry (RPDR) at MGB. The input, a csv/delimited file of labs for user-defined patients, may vary. Thus, users may need to adapt the initial data wrangling script based on the data input format. However, the downstream transformation, code-lab lookup tables, outcomes analysis, and LOINC remapping are standard for use with the provided REDCap Data Dictionary, DataDictionary_eLAB.csv. The available R-markdown ((https://github.com/TheMillerLab/eLAB) provides suggestions and instructions on where or when upfront script modifications may be necessary to accommodate input variability.

The eLAB pipeline takes several inputs. For example, the input for use with the ‘ehr_format(dt)’ single-line command is non-tabular data assigned as R object ‘dt’ with 4 columns: 1) Patient Name (MRN), 2) Collection Date, 3) Collection Time, and 4) Lab Results wherein several lab panels are in one data frame cell. A mock dataset in this ‘untidy-format’ is provided for demonstration purposes (https://github.com/TheMillerLab/eLAB).

Bulk lab data pulls often result in subtypes of the same lab. For example, potassium labs are reported as “Potassium,” “Potassium-External,” “Potassium(POC),” “Potassium,whole-bld,” “Potassium-Level-External,” “Potassium,venous,” and “Potassium-whole-bld/plasma.” eLAB utilizes a key-value lookup table with ~300 lab subtypes for remapping labs to the Data Dictionary (DD) code. eLAB reformats/accepts only those lab units pre-defined by the registry DD. The lab lookup table is provided for direct use or may be re-configured/updated to meet end-user specifications. eLAB is designed to remap, transform, and filter/adjust value units of semi-structured/structured bulk laboratory values data pulls from the EHR to align with the pre-defined code of the DD.

Data Dictionary (DD)

EHR clinical laboratory data is captured in REDCap using the ‘Labs’ repeating instrument (Supplemental Figures 1-2). The DD is provided for use by researchers at REDCap-participating institutions and is optimized to accommodate the same lab-type captured more than once on the same day for the same patient. The instrument captures 35 clinical lab types. The DD serves several major purposes in the eLAB pipeline. First, it defines every lab type of interest and associated lab unit of interest with a set field/variable name. It also restricts/defines the type of data allowed for entry for each data field, such as a string or numerics. The DD is uploaded into REDCap by every participating site/collaborator and ensures each site collects and codes the data the same way. Automation pipelines, such as eLAB, are designed to remap/clean and reformat data/units utilizing key-value look-up tables that filter and select only the labs/units of interest. eLAB ensures the data pulled from the EHR contains the correct unit and format pre-configured by the DD. The use of the same DD at every participating site ensures that the data field code, format, and relationships in the database are uniform across each site to allow for the simple aggregation of the multi-site data. For example, since every site in the MCCPR uses the same DD, aggregation is efficient and different site csv files are simply combined.

Study Cohort

This study was approved by the MGB IRB. Search of the EHR was performed to identify patients diagnosed with MCC between 1975-2021 (N=1,109) for inclusion in the MCCPR. Subjects diagnosed with primary cutaneous MCC between 2016-2019 (N= 176) were included in the test cohort for exploratory studies of lab result associations with overall survival (OS) using eLAB.

Statistical Analysis

OS is defined as the time from date of MCC diagnosis to date of death. Data was censored at the date of the last follow-up visit if no death event occurred. Univariable Cox proportional hazard modeling was performed among all lab predictors. Due to the hypothesis-generating nature of the work, p-values were exploratory and Bonferroni corrections were not applied.

Reporting of Aggregate Case and Death Count data was discontinued on May 11, 2023, with the expiration of the COVID-19 public health emergency declaration. Although these data will continue to be publicly available, this dataset will no longer be updated.

The surveillance case definition for COVID-19, a nationally notifiable disease, was first described in a position statement from the Council for State and Territorial Epidemiologists, which was later revised. However, there is some variation in how jurisdictions implemented these case definitions. More information on how CDC collects COVID-19 case surveillance data can be found at FAQ: COVID-19 Data and Surveillance.

Aggregate Data Collection Process Since the beginning of the COVID-19 pandemic, data were reported from state and local health departments through a robust process with the following steps:

• Aggregate county-level counts were obtained indirectly, via automated overnight web collection, or directly, via a data submission process.
• If more than one official county data source existed, CDC used a comprehensive data selection process comparing each official county data source to retrieve the highest case and death counts, unless otherwise specified by the state.
• A CDC data team reviewed counts for congruency prior to integration and set up alerts to monitor for discrepancies in the data.
• CDC routinely compiled these data and post the finalized information on COVID Data Tracker.
• County level data were aggregated to obtain state- and territory- specific totals.
• Counting of cases and deaths is based on date of report and not on the date of symptom onset. CDC calculates rates in these data by using population estimates provided by the US Census Bureau Population Estimates Program (2019 Vintage).
• COVID-19 aggregate case and death data are organized in a time series that includes cumulative number of cases and deaths as reported by a jurisdiction on a given date. New case and death counts are calculated as the week-to-week change in cumulative counts of cases and deaths reported (i.e., newly reported cases and deaths = cumulative number of cases/deaths reported this week minus the cumulative total reported the prior week.

This process was collaborative, with CDC and jurisdictions working together to ensure the accuracy of COVID-19 case and death numbers. County counts provided the most up-to-date numbers on cases and deaths by report date. Throughout data collection, CDC retrospectively updated counts to correct known data quality issues.

Description This archived public use dataset focuses on the cumulative and weekly case and death rates per 100,000 persons within various sociodemographic factors across all states and their counties. All resulting data are expressed as rates calculated as the number of cases or deaths per 100,000 persons in counties meeting various classification criteria using the US Census Bureau Population Estimates Program (2019 Vintage).

Each county within jurisdictions is classified into multiple categories for each factor. All rates in this dataset are based on classification of counties by the characteristics of their population, not individual-level factors. This applies to each of the available factors observed in this dataset. Specific factors and their corresponding categories are detailed below.

Population-level factors Each unique population factor is detailed below. Please note that the “Classification” column describes each of the 12 factors in the dataset, including a data dictionary describing what each numeric digit means within each classification. The “Category” column uses numeric digits (2-6, depending on the factor) defined in the “Classification” column.

Metro vs. Non-Metro – “Metro_Rural” Metro vs. Non-Metro classification type is an aggregation of the 6 National Center for Health Statistics (NCHS) Urban-Rural classifications, where “Metro” counties include Large Central Metro, Large Fringe Metro, Medium Metro, and Small Metro areas and “Non-Metro” counties include Micropolitan and Non-Core (Rural) areas. 1 – Metro, including “Large Central Metro, Large Fringe Metro, Medium Metro, and Small Metro” areas 2 – Non-Metro, including “Micropolitan, and Non-Core” areas

Urban/rural - “NCHS_Class” Urban/rural classification type is based on the 2013 National Center for Health Statistics Urban-Rural Classification Scheme for Counties. Levels consist of:

1 Large Central Metro
2 Large Fringe Metro 3 Medium Metro 4 Small Metro 5 Micropolitan 6 Non-Core (Rural)

American Community Survey (ACS) data were used to classify counties based on their age, race/ethnicity, household size, poverty level, and health insurance status distributions. Cut points were generated by using tertiles and categorized as High, Moderate, and Low percentages. The classification “Percent non-Hispanic, Native Hawaiian/Pacific Islander” is only available for “Hawaii” due to low numbers in this category for other available locations. This limitation also applies to other race/ethnicity categories within certain jurisdictions, where 0 counties fall into the certain category. The cut points for each ACS category are further detailed below:

Age 65 - “Age65”

1 Low (0-24.4%) 2 Moderate (>24.4%-28.6%) 3 High (>28.6%)

Non-Hispanic, Asian - “NHAA”

1 Low (<=5.7%) 2 Moderate (>5.7%-17.4%) 3 High (>17.4%)

Non-Hispanic, American Indian/Alaskan Native - “NHIA”

1 Low (<=0.7%) 2 Moderate (>0.7%-30.1%) 3 High (>30.1%)

Non-Hispanic, Black - “NHBA”

1 Low (<=2.5%) 2 Moderate (>2.5%-37%) 3 High (>37%)

Hispanic - “HISP”

1 Low (<=18.3%) 2 Moderate (>18.3%-45.5%) 3 High (>45.5%)

Population in Poverty - “Pov”

1 Low (0-12.3%) 2 Moderate (>12.3%-17.3%) 3 High (>17.3%)

Population Uninsured- “Unins”

1 Low (0-7.1%) 2 Moderate (>7.1%-11.4%) 3 High (>11.4%)

Average Household Size - “HH”

1 Low (1-2.4) 2 Moderate (>2.4-2.6) 3 High (>2.6)

Community Vulnerability Index Value - “CCVI” COVID-19 Community Vulnerability Index (CCVI) scores are from Surgo Ventures, which range from 0 to 1, were generated based on tertiles and categorized as:

1 Low Vulnerability (0.0-0.4) 2 Moderate Vulnerability (0.4-0.6) 3 High Vulnerability (0.6-1.0)

Social Vulnerability Index Value – “SVI" Social Vulnerability Index (SVI) scores (vintage 2020), which also range from 0 to 1, are from CDC/ASTDR’s Geospatial Research, Analysis & Service Program. Cut points for CCVI and SVI scores were generated based on tertiles and categorized as:

1 Low Vulnerability (0-0.333) 2 Moderate Vulnerability (0.334-0.666) 3 High Vulnerability (0.667-1)