TagX data annotation services are a set of tools and processes used to accurately label and classify large amounts of data for use in machine learning and artificial intelligence applications. The services are designed to be highly accurate, efficient, and customizable, allowing for a wide range of data types and use cases.

The process typically begins with a team of trained annotators reviewing and categorizing the data, using a variety of annotation tools and techniques, such as text classification, image annotation, and video annotation. The annotators may also use natural language processing and other advanced techniques to extract relevant information and context from the data.

Once the data has been annotated, it is then validated and checked for accuracy by a team of quality assurance specialists. Any errors or inconsistencies are corrected, and the data is then prepared for use in machine learning and AI models.

TagX annotation services can be applied to a wide range of data types, including text, images, videos, and audio. The services can be customized to meet the specific needs of each client, including the type of data, the level of annotation required, and the desired level of accuracy.

TagX data annotation services provide a powerful and efficient way to prepare large amounts of data for use in machine learning and AI applications, allowing organizations to extract valuable insights and improve their decision-making processes.

This is the accompanying data for the paper "Analyzing Dataset Annotation Quality Management in the Wild". Data quality is crucial for training accurate, unbiased, and trustworthy machine learning models and their correct evaluation. Recent works, however, have shown that even popular datasets used to train and evaluate state-of-the-art models contain a non-negligible amount of erroneous annotations, bias or annotation artifacts. There exist best practices and guidelines regarding annotation projects. But to the best of our knowledge, no large-scale analysis has been performed as of yet on how quality management is actually conducted when creating natural language datasets and whether these recommendations are followed. Therefore, we first survey and summarize recommended quality management practices for dataset creation as described in the literature and provide suggestions on how to apply them. Then, we compile a corpus of 591 scientific publications introducing text datasets and annotate it for quality-related aspects, such as annotator management, agreement, adjudication or data validation. Using these annotations, we then analyze how quality management is conducted in practice. We find that a majority of the annotated publications apply good or very good quality management. However, we deem the effort of 30% of the works as only subpar. Our analysis also shows common errors, especially with using inter-annotator agreement and computing annotation error rates.

Leaves from genetically unique Juglans regia plants were scanned using X-ray micro-computed tomography (microCT) on the X-ray μCT beamline (8.3.2) at the Advanced Light Source (ALS) in Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA USA). Soil samples were collected in Fall of 2017 from the riparian oak forest located at the Russell Ranch Sustainable Agricultural Institute at the University of California Davis. The soil was sieved through a 2 mm mesh and was air dried before imaging. A single soil aggregate was scanned at 23 keV using the 10x objective lens with a pixel resolution of 650 nanometers on beamline 8.3.2 at the ALS. Additionally, a drought stressed almond flower bud (Prunus dulcis) from a plant housed at the University of California, Davis, was scanned using a 4x lens with a pixel resolution of 1.72 µm on beamline 8.3.2 at the ALS Raw tomographic image data was reconstructed using TomoPy. Reconstructions were converted to 8-bit tif or png format using ImageJ or the PIL package in Python before further processing. Images were annotated using Intel’s Computer Vision Annotation Tool (CVAT) and ImageJ. Both CVAT and ImageJ are free to use and open source. Leaf images were annotated in following Théroux-Rancourt et al. (2020). Specifically, Hand labeling was done directly in ImageJ by drawing around each tissue; with 5 images annotated per leaf. Care was taken to cover a range of anatomical variation to help improve the generalizability of the models to other leaves. All slices were labeled by Dr. Mina Momayyezi and Fiona Duong.To annotate the flower bud and soil aggregate, images were imported into CVAT. The exterior border of the bud (i.e. bud scales) and flower were annotated in CVAT and exported as masks. Similarly, the exterior of the soil aggregate and particulate organic matter identified by eye were annotated in CVAT and exported as masks. To annotate air spaces in both the bud and soil aggregate, images were imported into ImageJ. A gaussian blur was applied to the image to decrease noise and then the air space was segmented using thresholding. After applying the threshold, the selected air space region was converted to a binary image with white representing the air space and black representing everything else. This binary image was overlaid upon the original image and the air space within the flower bud and aggregate was selected using the “free hand” tool. Air space outside of the region of interest for both image sets was eliminated. The quality of the air space annotation was then visually inspected for accuracy against the underlying original image; incomplete annotations were corrected using the brush or pencil tool to paint missing air space white and incorrectly identified air space black. Once the annotation was satisfactorily corrected, the binary image of the air space was saved. Finally, the annotations of the bud and flower or aggregate and organic matter were opened in ImageJ and the associated air space mask was overlaid on top of them forming a three-layer mask suitable for training the fully convolutional network. All labeling of the soil aggregate and soil aggregate images was done by Dr. Devin Rippner. These images and annotations are for training deep learning models to identify different constituents in leaves, almond buds, and soil aggregates Limitations: For the walnut leaves, some tissues (stomata, etc.) are not labeled and only represent a small portion of a full leaf. Similarly, both the almond bud and the aggregate represent just one single sample of each. The bud tissues are only divided up into buds scales, flower, and air space. Many other tissues remain unlabeled. For the soil aggregate annotated labels are done by eye with no actual chemical information. Therefore particulate organic matter identification may be incorrect. Resources in this dataset:Resource Title: Annotated X-ray CT images and masks of a Forest Soil Aggregate. File Name: forest_soil_images_masks_for_testing_training.zipResource Description: This aggregate was collected from the riparian oak forest at the Russell Ranch Sustainable Agricultural Facility. The aggreagate was scanned using X-ray micro-computed tomography (microCT) on the X-ray μCT beamline (8.3.2) at the Advanced Light Source (ALS) in Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA USA) using the 10x objective lens with a pixel resolution of 650 nanometers. For masks, the background has a value of 0,0,0; pores spaces have a value of 250,250, 250; mineral solids have a value= 128,0,0; and particulate organic matter has a value of = 000,128,000. These files were used for training a model to segment the forest soil aggregate and for testing the accuracy, precision, recall, and f1 score of the model.Resource Title: Annotated X-ray CT images and masks of an Almond bud (P. Dulcis). File Name: Almond_bud_tube_D_P6_training_testing_images_and_masks.zipResource Description: Drought stressed almond flower bud (Prunis dulcis) from a plant housed at the University of California, Davis, was scanned by X-ray micro-computed tomography (microCT) on the X-ray μCT beamline (8.3.2) at the Advanced Light Source (ALS) in Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA USA) using the 4x lens with a pixel resolution of 1.72 µm using. For masks, the background has a value of 0,0,0; air spaces have a value of 255,255, 255; bud scales have a value= 128,0,0; and flower tissues have a value of = 000,128,000. These files were used for training a model to segment the almond bud and for testing the accuracy, precision, recall, and f1 score of the model.Resource Software Recommended: Fiji (ImageJ),url: https://imagej.net/software/fiji/downloads Resource Title: Annotated X-ray CT images and masks of Walnut leaves (J. Regia) . File Name: 6_leaf_training_testing_images_and_masks_for_paper.zipResource Description: Stems were collected from genetically unique J. regia accessions at the 117 USDA-ARS-NCGR in Wolfskill Experimental Orchard, Winters, California USA to use as scion, and were grafted by Sierra Gold Nursery onto a commonly used commercial rootstock, RX1 (J. microcarpa × J. regia). We used a common rootstock to eliminate any own-root effects and to simulate conditions for a commercial walnut orchard setting, where rootstocks are commonly used. The grafted saplings were repotted and transferred to the Armstrong lathe house facility at the University of California, Davis in June 2019, and kept under natural light and temperature. Leaves from each accession and treatment were scanned using X-ray micro-computed tomography (microCT) on the X-ray μCT beamline (8.3.2) at the Advanced Light Source (ALS) in Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA USA) using the 10x objective lens with a pixel resolution of 650 nanometers. For masks, the background has a value of 170,170,170; Epidermis value= 85,85,85; Mesophyll value= 0,0,0; Bundle Sheath Extension value= 152,152,152; Vein value= 220,220,220; Air value = 255,255,255.Resource Software Recommended: Fiji (ImageJ),url: https://imagej.net/software/fiji/downloads

Data sets used for experimental evaluation in the related publication:Matching Web Tables with Knowledge Base Entities: From Entity Lookups to Entity EmbeddingsInternational Semantic Web Conference (1) 2017: 260-277Vasilis Efthymiou Oktie Hassanzadeh Mariano Rodríguez-Muro Vassilis ChristophidesThe gold standard data sets are collections of web tables:T2D (v1) consists of a schema-level gold standard of 1,748 Web tables, manually annotated with class- and property-mappings, as well as an entity-level gold standard of 233 Web tables.Limaye consists of 400 manually annotated Web tables with entity-, class-, and property-level correspondences, where single cells (not rows) are mapped to entities. The corrected version of this gold standard is adapted to annotate rows with entities, from the annotations of the label column cells.WikipediaGS is an instance-level gold standard developed from 485K Wikipedia tables, in which links in the label column are used to infer the annotation of a row to a DBpedia entity. Note on license: please refer to the README.txt. Data is derived from Wikipedia and other sources may have different licenses.Wikipedia contents can be shared under the terms of Creative Commons Attribution-ShareAlike Licenseas outlined on Wikipedia: https://en.wikipedia.org/wiki/Wikipedia:Reusing_Wikipedia_contentThe correspondences of the T2D Gold standard is provided under the terms of the Apache license. The Web tables are provided according the same terms of use, disclaimer of warranties and limitation of liabilities that apply to the Common Crawl corpus. The DBpedia subset is licensed under the terms of the Creative Commons Attribution-ShareAlike License and the GNU Free Documentation License that applies to DBpedia.Limaye gold standard is downloaded from: http://websail-fe.cs.northwestern.edu/TabEL/ (download date: August 25, 2016). Please refer to the original website and the following paper for more details and citation information:G. Limaye, S. Sarawagi, and S. Chakrabarti. Annotating and Searching Web Tables Using Entities, Types and Relationships. PVLDB, 3(1):1338–1347, 2010.THIS DATA IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

The Data Annotation and Collection Services market is experiencing robust growth, driven by the increasing adoption of artificial intelligence (AI) and machine learning (ML) across diverse sectors. The market, estimated at $10 billion in 2025, is projected to achieve a Compound Annual Growth Rate (CAGR) of 25% from 2025 to 2033, reaching approximately $45 billion by 2033. This significant expansion is fueled by several key factors. The surge in autonomous driving initiatives necessitates high-quality data annotation for training self-driving systems, while the burgeoning smart healthcare sector relies heavily on annotated medical images and data for accurate diagnoses and treatment planning. Similarly, the growth of smart security systems and financial risk control applications demands precise data annotation for improved accuracy and efficiency. Image annotation currently dominates the market, followed by text annotation, reflecting the widespread use of computer vision and natural language processing. However, video and voice annotation segments are showing rapid growth, driven by advancements in AI-powered video analytics and voice recognition technologies. Competition is intense, with both established technology giants like Alibaba Cloud and Baidu, and specialized data annotation companies like Appen and Scale Labs vying for market share. Geographic distribution shows a strong concentration in North America and Europe initially, but Asia-Pacific is expected to emerge as a major growth region in the coming years, driven primarily by China and India's expanding technology sectors. The market, however, faces certain challenges. The high cost of data annotation, particularly for complex tasks such as video annotation, can pose a barrier to entry for smaller companies. Ensuring data quality and accuracy remains a significant concern, requiring robust quality control mechanisms. Furthermore, ethical considerations surrounding data privacy and bias in algorithms require careful attention. To overcome these challenges, companies are investing in automation tools and techniques like synthetic data generation, alongside developing more sophisticated quality control measures. The future of the Data Annotation and Collection Services market will likely be shaped by advancements in AI and ML technologies, the increasing availability of diverse data sets, and the growing awareness of ethical considerations surrounding data usage.

Market Overview: The global data annotation service market is projected to reach a valuation of USD XXX million by 2033, expanding at a CAGR of XX% from 2025 to 2033. The surging demand for accurate and annotated data for artificial intelligence (AI) and machine learning (ML) models is driving the market growth. The increasing adoption of AI across various industries, including healthcare, manufacturing, and finance, is fueling the need for high-quality data annotation services. Market Dynamics and Key Players: Key drivers of the data annotation service market include the growing demand for automated processes, the rise of IoT devices generating massive data, and advancements in AI technology. However, the high cost of data annotation and concerns over data privacy pose challenges. The market is segmented into application areas (government, enterprise, others) and annotation types (text, image, others). Notable companies operating in the market include Appen Limited, CloudApp, Cogito Tech LLC, and Deep Systems. Regional markets include North America, Europe, Asia Pacific, and the Middle East & Africa. The study period spans from 2019 to 2033, with 2025 as the base year and a forecast period from 2025 to 2033.

The Asia Pacific data annotation tools market is projected to exhibit a robust CAGR of 28.05% during the forecast period of 2025-2033. This growth is primarily driven by the surging demand for high-quality annotated data for training and developing artificial intelligence (AI) and machine learning (ML) algorithms. The increasing adoption of AI and ML across various industry verticals, such as healthcare, retail, and financial services, is fueling the need for accurate and reliable data annotation. Key trends influencing the market growth include the rise of self-supervised annotation techniques, advancements in natural language processing (NLP), and the proliferation of cloud-based annotation platforms. Additionally, the growing awareness of the importance of data privacy and security is driving the adoption of annotation tools that comply with industry regulations. The competitive landscape features a mix of established players and emerging startups offering a wide range of annotation tools. The Asia Pacific data annotation tools market is projected to grow from USD 2.4 billion in 2022 to USD 10.5 billion by 2027, at a CAGR of 35.4% during the forecast period. The growth of the market is attributed to the increasing adoption of artificial intelligence (AI) and machine learning (ML) technologies, which require large amounts of annotated data for training and development.

The global data annotation platform market is experiencing robust growth, driven by the increasing demand for high-quality training data across diverse sectors. The market's expansion is fueled by the proliferation of artificial intelligence (AI) and machine learning (ML) applications in autonomous driving, smart healthcare, and financial risk control. Autonomous vehicles, for instance, require vast amounts of annotated data for object recognition and navigation, significantly boosting demand. Similarly, the healthcare sector leverages data annotation for medical image analysis, leading to advancements in diagnostics and treatment. The market is segmented by application (Autonomous Driving, Smart Healthcare, Smart Security, Financial Risk Control, Social Media, Others) and annotation type (Image, Text, Voice, Video, Others). The prevalent use of cloud-based platforms, coupled with the rising adoption of AI across various industries, presents significant opportunities for market expansion. While the market faces challenges such as high annotation costs and data privacy concerns, the overall growth trajectory remains positive, with a projected compound annual growth rate (CAGR) suggesting substantial market expansion over the forecast period (2025-2033). Competition among established players like Appen, Amazon, and Google, alongside emerging players focusing on specialized annotation needs, is expected to intensify. The regional distribution of the market reflects the concentration of AI and technology development in specific geographical regions. North America and Europe currently hold a significant market share due to their robust technological infrastructure and early adoption of AI technologies. However, the Asia-Pacific region, particularly China and India, is demonstrating rapid growth potential due to the burgeoning AI industry and expanding digital economy. This signifies a shift in market dynamics, as the demand for data annotation services increases globally, leading to a more geographically diverse market landscape. Continuous advancements in annotation techniques, including the use of automated tools and crowdsourcing, are expected to reduce costs and improve efficiency, further fueling market growth.

The global data annotation tools market is anticipated to grow significantly over the forecast period, reaching a projected value of 1,639.44 million by 2033. This growth is attributed to the rising demand for data annotation in the fields of artificial intelligence (AI), machine learning (ML), and data science. The increase in the volume and complexity of data being generated is also contributing to the market growth. Key drivers of the market include the increasing adoption of AI and ML across various industries, the need for accurate data annotation for training machine learning models, and the growing demand for data annotation services for applications such as object detection, image segmentation, and natural language processing. Some of the major players in the market include IBM, Google, Microsoft, Amazon Web Services (AWS), and Hive. Key drivers for this market are: AI and ML advancementsExpansion of autonomous vehiclesGrowth of smart citiesProliferation of IoT devicesRise of cloud computing. Potential restraints include: Growing adoption of AI and MLIncreasing demand for high-quality annotated dataRise of data-intensive applicationsEmergence of cloud-based annotation toolsGrowing need for data governance and compliance.

This data presents fire segmentation annotation data on 12 commonly used and publicly available “VisiFire Dataset” videos from http://signal.ee.bilkent.edu.tr/VisiFire/. This annotations dataset was obtained by per-frame, manual hand annotation over the fire region with 2,684 total annotated frames. Since this annotation provides per-frame segmentation data, it offers a new and unique fire motion feature to the existing video, unlike other fire segmentation data that are collected from different still images. The annotations dataset also provides ground truth for segmentation task on videos. With segmentation task, it offers better insight on how well a machine learning model understood, not only detecting whether a fire is present, but also its exact location by calculating metrics such as Intersection over Union (IoU) with this annotations data. This annotations data is a tremendously useful addition to train, develop, and create a much better smart surveillance system for early detection in high-risk fire hotspots area.

md5sum: 50e63a6ee394751fad75dc43017710e8 QUT Research Data Respository Dataset Resource available for download

The market for AI-assisted annotation tools is projected to experience significant growth in the coming years, driven by the increasing adoption of machine learning, computer vision, and artificial intelligence technologies. The market is expected to reach a value of 617 million USD by 2033, growing at a CAGR of 9.2%. This growth is attributed to the increasing demand for high-quality annotated data for training AI models and the growing adoption of AI-powered solutions across various industries. Key drivers of the market include the increasing adoption of machine learning and deep learning technologies, the growing demand for high-quality annotated data, and the increasing adoption of AI-powered solutions across various industries. Some major trends include the increasing adoption of cloud-based AI-assisted annotation tools, the growing use of AI-assisted annotation tools for video and audio data, and the increasing use of AI-assisted annotation tools for real-time applications. Key restraints include the high cost of AI-assisted annotation tools, the lack of skilled professionals, and the ethical concerns associated with using AI for annotation. Key segments include application, type, and region. Major companies operating in the market include NVIDIA, DataGym, Dataloop, Encord, Hive Data, IBM Watson Studio, Innodata, LabelMe, Scale AI, SuperAnnotate, Supervisely, V7, and VoTT. The market is expected to be dominated by North America, followed by Europe and Asia Pacific.

Shoplifting Annotation

## Overview

Shoplifting Annotation is a dataset for object detection tasks - it contains Shopping Containers annotations for 768 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

NB: In order to reproduce figures found in the article in the simplest manner, one should clone the GitHub repository and run the plotting script directly after installing necessary requirements. The corresponding files used for making these plots can also be found here under "output.zip". Classifying tweets with large language models with zero- and few-shot learning with custom and generic prompts, as well as supervised learning algorithms for comparison. The full GitHub repository for this data can be found at this URL or by following the link under the "metadata" tab. This GitHub repo contains an extensive README file explaining how to run the code and reproduce the results and plots found in the article. The present Dataverse repository contains all code and prompts used to generate predictions on the human annotated data, as well as the code book used by human annotators on this data. Due to data sharing policies at X (formerly Twitter), we are unable to share full texts from the Tweets used in our study. Instead, we have provided Tweet IDs, unique identifiers for all 2000 tweets used in our experiments (1000 of each annotation type) which can be used to re-scrape the data if desired.

84,516 Sentences - English Intention Annotation Data in Interactive Scenes, annotated with intent classes, including slot and slot value information; the intent field includes music, weather, date, schedule, home equipment, etc.; it is applied to intent recognition research and related fields.

Introduction: The goal of this dataset is to aid NLP research on recognizing safety critical information from drug usage guideline or patient handout data. This dataset contains annotated advice statements from 90 online DUG documents that corresponds to 90 drugs or medications that are used in the prescriptions of patients suffering from one or more chronic diseases. The advice statements are annotated in eight safety-critical categories: activity or lifestyle related, disease or symptom related, drug administration related, exercise related, food or beverage related, other drug related, pregnancy related, and temporal.

Data Collection: The data was collected from MedScape. It is one of the most widely used reference for health care providers. At first, 34 real anonymized prescriptions of patients suffering from one or more chronic diseases are collected. These prescriptions contains 165 drugs that are used to treat chronic diseases. Then, MedScape was crawled to collect the drug user guideline (DUG) / patient handout for these 165 drugs. But, MedScape does not have DUG document for all drugs. We found DUG document for 90 drugs in MedScape.

Data Annotation tool: The data annotation tool is developed to ease the annotation process. It allows the user to select a DUG document and select a position from the document in terms of line number. It stores the user log from the annotator and loads the most recent position from the log when the application is launched. It supports annotating multiple files for the same drug, as often there are multiple overlapping sources of drug usage guidelines for a single drug. Often DUG documents contain formatted text. This tool aids annotation of the formatted text as well. The annotation tool is also available upon request.

Annotated Data Description: The annotated data contains the annotation tag(s) of each advice extracted from the 90 online DUG documents. It also contains the phrases or topics in the advice statement that triggers the annotation tag, such as, activity, exercise, medication name, food or beverage name, disease name, pregnancy condition (gestational, postpartum). Sometimes disease names are not directly mentioned rather mentioned as a condition (e.g., stomach bleeding, alcohol abuse) or state of a parameter (e.g., low blood sugar, low blood pressure). The annotated data is formatted as following:
drug name, drug number, line number of the first sentence of the advice in the DUG document, advice Text, advice tag(s), medication, food, activity, exercise, and disease names mentioned in the advice.

Unannotated Data Description:
The unannotated data contains the raw DUG document for 90 drugs. It also contains the drug interaction information for the 165 drugs. The drug interaction information is categorized in 4 classes, contraindicated, serious, monitor closely, and minor. This information can be utilized to automatically detect potential interaction and effect of interaction among multiple drugs.

Citation: If you use this dataset in your work, please cite the following reference in any publication:

@inproceedings{preum2018DUG,
title={A Corpus of Drug Usage Guidelines Annotated with Type of Advice},
author={Sarah Masud Preum, Md. Rizwan Parvez, Kai-Wei Chang, and John A. Stankovic},
booktitle={ Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)},
publisher = {European Language Resources Association (ELRA)},
year={2018}
}

The POLIANNA dataset is a collection of legislative texts from the European Union (EU) that have been annotated based on theoretical concepts of policy design. The dataset consists of 20,577 annotated spans in 412 articles, drawn from 18 EU climate change mitigation and renewable energy laws, and can be used to develop supervised machine learning approaches for scaling policy analysis. The dataset includes a novel coding scheme for annotating text spans, and you find a description of the annotated corpus, an analysis of inter-annotator agreement, and a discussion of potential applications in the paper accompanying this dataset. The objective of this dataset to build tools that assist with manual coding of policy texts by automatically identifying relevant paragraphs.

Detailed instructions and further guidance about the dataset as well as all the code used for this project can be found in the accompanying paper and on the GitHub project page. The repository also contains useful code to calculate various inter-annotator agreement measures and can be used to process text annotations generated by INCEpTION.

Dataset Description

We provide the dataset in 3 different formats:JSON: Each article corresponds to a folder, where the Tokens and Spans are stored in a separate JSON file. Each article-folder further contains the raw policy-text as in a text file and the metadata about the policy. This is the most human-readable format.

JSONL: Same folder structure as the JSON format, but the Spans and Tokens are stored in a JSONL file, where each line is a valid JSON document.

Pickle: We provide the dataset as a Python object. This is the recommended method when using our own Python framework that is provided on GitHub. For more information, check out the GitHub project page.

License

The POLIANNA dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. If you use the POLIANNA dataset in your research in any form, please cite the dataset.

Citation

Sewerin, S., Kaack, L.H., Küttel, J. et al. Towards understanding policy design through text-as-data approaches: The policy design annotations (POLIANNA) dataset. Sci Data10, 896 (2023). https://doi.org/10.1038/s41597-023-02801-z

Annotation studies often require annotators to familiarize themselves with the task, its annotation scheme, and the data domain. This can be overwhelming in the beginning, mentally taxing, and induce errors into the resulting annotations; especially in citizen science or crowd sourcing scenarios where domain expertise is not required and only annotation guidelines are provided. To alleviate these issues, we propose annotation curricula, a novel approach to implicitly train annotators. We gradually introduce annotators into the task by ordering instances that are annotated according to a learning curriculum. To do so, we first formalize annotation curricula for sentence- and paragraph-level annotation tasks, define an ordering strategy, and identify well-performing heuristics and interactively trained models on three existing English datasets. We then conduct a user study with 40 voluntary participants who are asked to identify the most fitting misconception for English tweets about the Covid-19 pandemic. Our results show that using a simple heuristic to order instances can already significantly reduce the total annotation time while preserving a high annotation quality. Annotation curricula thus can provide a novel way to improve data collection. To facilitate future research, we further share our code and data consisting of 2,400 annotations.

If you use this dataset, please cite this paper: Puertas, E.; De-Las-Heras, G.; Sánchez-Soriano, J.; Fernández-Andrés, J. Dataset: Variable Message Signal Annotated Images for Object Detection. Data 2022, 7, 41. https://doi.org/10.3390/data7040041

This dataset consists of Spanish road images taken from inside a vehicle, as well as annotations in XML files in PASCAL VOC format that indicate the location of Variable Message Signals within them. Also, a CSV file is attached with information regarding the geographic position, the folder where the image is located, and the text in Spanish. This can be used to train supervised learning computer vision algorithms, such as convolutional neural networks. Throughout this work, the process followed to obtain the dataset, image acquisition, and labeling, and its specifications are detailed. The dataset is constituted of 1216 instances, 888 positives, and 328 negatives, in 1152 jpg images with a resolution of 1280x720 pixels. These are divided into 576 real images and 576 images created from the data-augmentation technique. The purpose of this dataset is to help in road computer vision research since there is not one specifically for VMSs.

The folder structure of the dataset is as follows:

• vms_dataset/
  • data.csv
  • real_images/
    • imgs/
    • annotations/
  • data-augmentation/
    • imgs/
    • annotations/

In which:

• data.csv: Each row contains the following information separated by commas (,): image_name, x_min, y_min, x_max, y_max, class_name, lat, long, folder, text.
• real_images: Images extracted directly from the videos.
• data-augmentation: Images created using data-augmentation
• imgs: Image files in .jpg format.
• annotations: Annotation files in .xml format.

Researcher(s): Alexandros Mokas, Eleni Kamateri

Supervisor: Ioannis Tsampoulatidis

This repository contains 3 social media datasets:

2 Post-processing datasets: These datasets contain post-processing data extracted from the analysis of social media posts collected for two different use cases during the first two years of the Deepcube project. More specifically, these include:

• The UC2 dataset containing the post-processing analysis of the Twitter data collected for the DeepCube use case (UC2) dealing with the climate induced migration in Africa. This dataset contains in total 5,695,253 social media posts collected from the Twitter platform, based on the initial version of search criteria relevant to UC2 defined by Universitat De Valencia, focused on the regions of Ethiopia and Somalia and started from 26 June, 2021 till March, 2023.
• The UC5 dataset containing the post-processing analysis of the Twitter and Instagram data collected for the DeepCube use case (UC5) related to the sustainable and environmentally-friendly tourism. This dataset contains in total 58,143 social media posts collected from the Twitter and Instagram platform (12,881 collected from Twitter and 45,262 collected from Instagram), based on the initial version of search criteria relevant to UC5 defined by MURMURATION SAS, focused on the regions of Brasil and started from 26 June, 2021 till March, 2023.

1 Annotated dataset: An additional anottated dataset was created that contains post-processing data along with annotations of Twitter posts collected for UC2 for the years 2010-2022. More specifically, it includes:

• The UC2 dataset contain the post-processing of the Twitter data collected for the DeepCube use case (UC2) dealing with the climate induced migration in Africa. This dataset contains in total 1721 annotated (412 relevant and 1309 irrelevant) by social media posts collected from the Twitter platform, focused on the region of Somalia and started from 1 January, 2010 till 31 December, 2022.

For every social media post retrieved from Twitter and Instagram, a preprocessing step was performed. This involved a three-step analysis of each post using the appropriate web service. First, the location of the post was automatically extracted from the text using a location extraction service. Second, the images included in the post were analyzed using a concept extraction service, which identified and provided the top ten concepts that best described the image. These concepts included items such as "person," "building," "drought," "sun," and so on. Finally, the sentiment expressed in the post's text was determined by using a sentiment analysis service. The sentiment was classified as either positive, negative, or neutral.

After the social media posts were preprocessed, they were visualized using the Social Media Web Application. This intuitive, user-friendly online application was designed for both expert and non-expert users and offers a web-based user interface for filtering and visualizing the collected social media data. The application provides various filtering options, an interactive map, a timeline, and a collection of graphs to help users analyze the data. Moreover, this application provides users with the option to download aggregated data for specific periods by applying filters and clicking the "Download Posts" button. This feature allows users to easily extract and analyze social media data outside of the web application, providing greater flexibility and control over data analysis.

The dataset is provided by INFALIA.

INFALIA, being a spin-off of the CERTH institute and a partner of a research EU project, releases this dataset containing Tweets IDs and post pre-processing data for the sole purpose of enabling the validation of the research conducted within the DeepCube. Moreover, Twitter Content provided in this dataset to third parties remains subject to the Twitter Policy, and those third parties must agree to the Twitter Terms of Service, Privacy Policy, Developer Agreement, and Developer Policy (https://developer.twitter.com/en/developer-terms) before receiving this download.