This dataset is a compilation of processed data on citation and references for research papers including their author, institution and open access info for a selected sample of academics analysed using Microsoft Academic Graph (MAG) data and CORE. The data for this dataset was collected during December 2019 to January 2020.Six countries (Austria, Brazil, Germany, India, Portugal, United Kingdom and United States) were the focus of the six questions which make up this dataset. There is one csv file per country and per question (36 files in total). More details about the creation of this dataset are available on the public ON-MERRIT D3.1 deliverable report.The dataset is a combination of two different data sources, one part is a dataset created on analysing promotion policies across the target countries, while the second part is a set of data points available to understand the publishing behaviour. To facilitate the analysis the dataset is organised in the following seven folders:PRTThe dataset with the file name "PRT_policies.csv" contains the related information as this was extracted from promotion, review and tenure (PRT) policies. Q1: What % of papers coming from a university are Open Access?- Dataset Name format: oa_status_countryname_papers.csv- Dataset Contents: Open Access (OA) status of all papers of all the universities listed in Times Higher Education World University Rankings (THEWUR) for the given country. A paper is marked OA if there is at least an OA link available. OA links are collected using the CORE Discovery API.- Important considerations about this dataset: - Papers with multiple authorship are preserved only once towards each of the distinct institutions their authors may belong to. - The service we used to recognise if a paper is OA, CORE Discovery, does not contain entries for all paperids in MAG. This implies that some of the records in the dataset extracted will not have either a true or false value for the _is_OA_ field. - Only those records marked as true for _is_OA_ field can be said to be OA. Others with false or no value for is_OA field are unknown status (i.e. not necessarily closed access).Q2: How are papers, published by the selected universities, distributed across the three scientific disciplines of our choice?- Dataset Name format: fsid_countryname_papers.csv- Dataset Contents: For the given country, all papers for all the universities listed in THEWUR with the information of fieldofstudy they belong to.- Important considerations about this dataset: * MAG can associate a paper to multiple fieldofstudyid. If a paper belongs to more than one of our fieldofstudyid, separate records were created for the paper with each of those _fieldofstudyid_s.- MAG assigns fieldofstudyid to every paper with a score. We preserve only those records whose score is more than 0.5 for any fieldofstudyid it belongs to.- Papers with multiple authorship are preserved only once towards each of the distinct institutions their authors may belong to. Papers with authorship from multiple universities are counted once towards each of the universities concerned.Q3: What is the gender distribution in authorship of papers published by the universities?- Dataset Name format: author_gender_countryname_papers.csv- Dataset Contents: All papers with their author names for all the universities listed in THEWUR.- Important considerations about this dataset :- When there are multiple collaborators(authors) for the same paper, this dataset makes sure that only the records for collaborators from within selected universities are preserved.- An external script was executed to determine the gender of the authors. The script is available here.Q4: Distribution of staff seniority (= number of years from their first publication until the last publication) in the given university.- Dataset Name format: author_ids_countryname_papers.csv- Dataset Contents: For a given country, all papers for authors with their publication year for all the universities listed in THEWUR.- Important considerations about this work :- When there are multiple collaborators(authors) for the same paper, this dataset makes sure that only the records for collaborators from within selected universities are preserved.- Calculating staff seniority can be achieved in various ways. The most straightforward option is to calculate it as _academic_age = MAX(year) - MIN(year) _for each authorid.Q5: Citation counts (incoming) for OA vs Non-OA papers published by the university.- Dataset Name format: cc_oa_countryname_papers.csv- Dataset Contents: OA status and OA links for all papers of all the universities listed in THEWUR and for each of those papers, count of incoming citations available in MAG.- Important considerations about this dataset :- CORE Discovery was used to establish the OA status of papers.- Papers with multiple authorship are preserved only once towards each of the distinct institutions their authors may belong to.- Only those records marked as true for _is_OA_ field can be said to be OA. Others with false or no value for is_OA field are unknown status (i.e. not necessarily closed access).Q6: Count of OA vs Non-OA references (outgoing) for all papers published by universities.- Dataset Name format: rc_oa_countryname_-papers.csv- Dataset Contents: Counts of all OA and unknown papers referenced by all papers published by all the universities listed in THEWUR.- Important considerations about this dataset :- CORE Discovery was used to establish the OA status of papers being referenced.- Papers with multiple authorship are preserved only once towards each of the distinct institutions their authors may belong to. Papers with authorship from multiple universities are counted once towards each of the universities concerned.Additional files:- _fieldsofstudy_mag_.csv: this file contains a dump of fieldsofstudy table of MAG mapping each of the ids to their actual field of study name.

This dataset contains the results of a literature review of experimental nutrient addition studies to determine which nutrient forms were most often measured in the scientific literature. To obtain a representative selection of relevant studies, we searched Web of Science™ using a search string to target experimental studies in artificial and natural lotic systems while limiting irrelevant papers. We screened the titles and abstracts of returned papers for relevance (experimental studies in streams/stream mesocosms that manipulated nutrients). To supplement this search, we sorted the relevant articles from the Web of Science™ search alphabetically by author and sequentially examined the bibliographies for additional relevant articles (screening titles for relevance, and then screening abstracts of potentially relevant articles) until we had obtained a total of 100 articles. If we could not find a relevant article electronically, we moved to the next article in the bibliography. Our goal was not to be completely comprehensive, but to obtain a fairly large sample of published, peer-reviewed studies from which to assess patterns. We excluded any lentic or estuarine studies from consideration and included only studies that used mesocosms mimicking stream systems (flowing water or stream water source) or that manipulated nutrient concentrations in natural streams or rivers. We excluded studies that used nutrient diffusing substrate (NDS) because these manipulate nutrients on substrates and not in the water column. We also excluded studies examining only nutrient uptake, which rely on measuring dissolved nutrient concentrations with the goal of characterizing in-stream processing (e.g., Newbold et al., 1983). From the included studies, we extracted or summarized the following information: study type, study duration, nutrient treatments, nutrients measured, inclusion of TN and/or TP response to nutrient additions, and a description of how results were reported in relation to the research-management mismatch, if it existed. Below is information on how the search was conducted: Search string used for Web of Science advanced search Search conducted on 27 September 2016. TS= (stream OR creek OR river* OR lotic OR brook OR headwater OR tributary) AND TS = (mesocosm OR flume OR "artificial stream" OR "experimental stream" OR "nutrient addition") AND TI= (nitrogen OR phosphorus OR nutrient OR enrichment OR fertilization OR eutrophication)

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

Background: Attribution to the original contributor upon reuse of published data is important both as a reward for data creators and to document the provenance of research findings. Previous studies have found that papers with publicly available datasets receive a higher number of citations than similar studies without available data. However, few previous analyses have had the statistical power to control for the many variables known to predict citation rate, which has led to uncertain estimates of the "citation benefit". Furthermore, little is known about patterns in data reuse over time and across datasets. Method and Results: Here, we look at citation rates while controlling for many known citation predictors, and investigate the variability of data reuse. In a multivariate regression on 10,555 studies that created gene expression microarray data, we found that studies that made data available in a public repository received 9% (95% confidence interval: 5% to 13%) more citations than similar studies for which the data was not made available. Date of publication, journal impact factor, open access status, number of authors, first and last author publication history, corresponding author country, institution citation history, and study topic were included as covariates. The citation benefit varied with date of dataset deposition: a citation benefit was most clear for papers published in 2004 and 2005, at about 30%. Authors published most papers using their own datasets within two years of their first publication on the dataset, whereas data reuse papers published by third-party investigators continued to accumulate for at least six years. To study patterns of data reuse directly, we compiled 9,724 instances of third party data reuse via mention of GEO or ArrayExpress accession numbers in the full text of papers. The level of third-party data use was high: for 100 datasets deposited in year 0, we estimated that 40 papers in PubMed reused a dataset by year 2, 100 by year 4, and more than 150 data reuse papers had been published by year 5. Data reuse was distributed across a broad base of datasets: a very conservative estimate found that 20% of the datasets deposited between 2003 and 2007 had been reused at least once by third parties. Conclusion: After accounting for other factors affecting citation rate, we find a robust citation benefit from open data, although a smaller one than previously reported. We conclude there is a direct effect of third-party data reuse that persists for years beyond the time when researchers have published most of the papers reusing their own data. Other factors that may also contribute to the citation benefit are considered.We further conclude that, at least for gene expression microarray data, a substantial fraction of archived datasets are reused, and that the intensity of dataset reuse has been steadily increasing since 2003.

Self-citation analysis data based on PubMed Central subset (2002-2005) ---------------------------------------------------------------------- Created by Shubhanshu Mishra, Brent D. Fegley, Jana Diesner, and Vetle Torvik on April 5th, 2018 ## Introduction This is a dataset created as part of the publication titled: Mishra S, Fegley BD, Diesner J, Torvik VI (2018) Self-Citation is the Hallmark of Productive Authors, of Any Gender. PLOS ONE. It contains files for running the self citation analysis on articles published in PubMed Central between 2002 and 2005, collected in 2015. The dataset is distributed in the form of the following tab separated text files: * Training_data_2002_2005_pmc_pair_First.txt (1.2G) - Data for first authors * Training_data_2002_2005_pmc_pair_Last.txt (1.2G) - Data for last authors * Training_data_2002_2005_pmc_pair_Middle_2nd.txt (964M) - Data for middle 2nd authors * Training_data_2002_2005_pmc_pair_txt.header.txt - Header for the data * COLUMNS_DESC.txt file - Descriptions of all columns * model_text_files.tar.gz - Text files containing model coefficients and scores for model selection. * results_all_model.tar.gz - Model coefficient and result files in numpy format used for plotting purposes. v4.reviewer contains models for analysis done after reviewer comments. * README.txt file ## Dataset creation Our experiments relied on data from multiple sources including properitery data from Thompson Rueter's (now Clarivate Analytics) Web of Science collection of MEDLINE citations. Author's interested in reproducing our experiments should personally request from Clarivate Analytics for this data. However, we do make a similar but open dataset based on citations from PubMed Central which can be utilized to get similar results to those reported in our analysis. Furthermore, we have also freely shared our datasets which can be used along with the citation datasets from Clarivate Analytics, to re-create the datased used in our experiments. These datasets are listed below. If you wish to use any of those datasets please make sure you cite both the dataset as well as the paper introducing the dataset. * MEDLINE 2015 baseline: https://www.nlm.nih.gov/bsd/licensee/2015_stats/baseline_doc.html * Citation data from PubMed Central (original paper includes additional citations from Web of Science) * Author-ity 2009 dataset: - Dataset citation: Torvik, Vetle I.; Smalheiser, Neil R. (2018): Author-ity 2009 - PubMed author name disambiguated dataset. University of Illinois at Urbana-Champaign. https://doi.org/10.13012/B2IDB-4222651_V1 - Paper citation: Torvik, V. I., & Smalheiser, N. R. (2009). Author name disambiguation in MEDLINE. ACM Transactions on Knowledge Discovery from Data, 3(3), 1–29. https://doi.org/10.1145/1552303.1552304 - Paper citation: Torvik, V. I., Weeber, M., Swanson, D. R., & Smalheiser, N. R. (2004). A probabilistic similarity metric for Medline records: A model for author name disambiguation. Journal of the American Society for Information Science and Technology, 56(2), 140–158. https://doi.org/10.1002/asi.20105 * Genni 2.0 + Ethnea for identifying author gender and ethnicity: - Dataset citation: Torvik, Vetle (2018): Genni + Ethnea for the Author-ity 2009 dataset. University of Illinois at Urbana-Champaign. https://doi.org/10.13012/B2IDB-9087546_V1 - Paper citation: Smith, B. N., Singh, M., & Torvik, V. I. (2013). A search engine approach to estimating temporal changes in gender orientation of first names. In Proceedings of the 13th ACM/IEEE-CS joint conference on Digital libraries - JCDL ’13. ACM Press. https://doi.org/10.1145/2467696.2467720 - Paper citation: Torvik VI, Agarwal S. Ethnea -- an instance-based ethnicity classifier based on geo-coded author names in a large-scale bibliographic database. International Symposium on Science of Science March 22-23, 2016 - Library of Congress, Washington DC, USA. http://hdl.handle.net/2142/88927 * MapAffil for identifying article country of affiliation: - Dataset citation: Torvik, Vetle I. (2018): MapAffil 2016 dataset -- PubMed author affiliations mapped to cities and their geocodes worldwide. University of Illinois at Urbana-Champaign. https://doi.org/10.13012/B2IDB-4354331_V1 - Paper citation: Torvik VI. MapAffil: A Bibliographic Tool for Mapping Author Affiliation Strings to Cities and Their Geocodes Worldwide. D-Lib magazine : the magazine of the Digital Library Forum. 2015;21(11-12):10.1045/november2015-torvik * IMPLICIT journal similarity: - Dataset citation: Torvik, Vetle (2018): Author-implicit journal, MeSH, title-word, and affiliation-word pairs based on Author-ity 2009. University of Illinois at Urbana-Champaign. https://doi.org/10.13012/B2IDB-4742014_V1 * Novelty dataset for identify article level novelty: - Dataset citation: Mishra, Shubhanshu; Torvik, Vetle I. (2018): Conceptual novelty scores for PubMed articles. University of Illinois at Urbana-Champaign. https://doi.org/10.13012/B2IDB-5060298_V1 - Paper citation: Mishra S, Torvik VI. Quantifying Conceptual Novelty in the Biomedical Literature. D-Lib magazine : The Magazine of the Digital Library Forum. 2016;22(9-10):10.1045/september2016-mishra - Code: https://github.com/napsternxg/Novelty * Expertise dataset for identifying author expertise on articles: * Source code provided at: https://github.com/napsternxg/PubMed_SelfCitationAnalysis Note: The dataset is based on a snapshot of PubMed (which includes Medline and PubMed-not-Medline records) taken in the first week of October, 2016. Check here for information to get PubMed/MEDLINE, and NLMs data Terms and Conditions Additional data related updates can be found at Torvik Research Group ## Acknowledgments This work was made possible in part with funding to VIT from NIH grant P01AG039347 and NSF grant 1348742. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. ## License Self-citation analysis data based on PubMed Central subset (2002-2005) by Shubhanshu Mishra, Brent D. Fegley, Jana Diesner, and Vetle Torvik is licensed under a Creative Commons Attribution 4.0 International License. Permissions beyond the scope of this license may be available at https://github.com/napsternxg/PubMed_SelfCitationAnalysis.

This record contains the datasets and models used and produced for the work reported in the paper "Combining Visual and Textual Features for Semantic Segmentation of Historical Newspapers" (link).

Please cite this paper if you are using the models/datasets or find it relevant to your research:

@article{barman_combining_2020,
  title = {{Combining Visual and Textual Features for Semantic Segmentation of Historical Newspapers}},
  author = {Raphaël Barman and Maud Ehrmann and Simon Clematide and Sofia Ares Oliveira and Frédéric Kaplan},
  journal= {Journal of Data Mining \& Digital Humanities},
  volume= {HistoInformatics}
  DOI = {10.5281/zenodo.4065271},
  year = {2021},
  url = {https://jdmdh.episciences.org/7097},
}

Please note that this record contains data under different licenses.

1. DATA

• Annotations (json files): JSON files contains image annotations, with one file per newspaper containing region annotations (label and coordinates) in VIA format. The following licenses apply:
  • luxwort.json: those annotations are under a CC0 1.0 license. Please refer to the right statement specified for each image in the file.
  • GDL.json, IMP.json and JDG.json: those annotations are under a CC BY-SA 4.0 license.

• Image files: The archive images.zip contains the Swiss titles image files (GDL, IMP, JDG) used for the experiments described in the paper. Those images are under copyright (property of the journal Le Temps and of ArcInfo) and can be used for academic research or educational purposes only. Redistribution, publication or commercial use are not permitted. These terms of use are similar to the following right statement: http://rightsstatements.org/vocab/InC-EDU/1.0/

2. MODELS

Some of the best models are released under a CC BY-SA 4.0 license (they are also available as assets of the current Github release).

• JDG_flair-FT: this model was trained on JDG using french Flair and FastText embeddings. It is able to predict the four classes presented in the paper (Serial, Weather, Death notice and Stocks).
• Luxwort_obituary_flair-bpemb: this model was trained on Luxwort using multilingual Flair and Byte-pair embeddings. It is able to predict the Death notice class.
• Luxwort_obituary_flair-FT_indomain: this model was trained on Luxwort using in-domain Flair and FastText embeddings (trained on Luxwort data). It is also able to predict the Death notice class.

Those models can be used to predict probabilities on new images using the same code as in the original dhSegment repository. One needs to adjust three parameters to the predict function: 1) embeddings_path (the path to the embeddings list), 2) embeddings_map_path(the path to the compressed embedding map), and 3) embeddings_dim (the size of the embeddings).

Please refer to the paper for further information or contact us.

3. CODE:

https://github.com/dhlab-epfl/dhSegment-text

4. ACKNOWLEDGEMENTS
We warmly thank the journal Le Temps (owner of La Gazette de Lausanne and the Journal de Genève) and the group ArcInfo (owner of L'Impartial) for accepting to share the related datasets for academic purposes. We also thank the National Library of Luxembourg for its support with all steps related to the Luxemburger Wort annotation release.
This work was realized in the context of the impresso - Media Monitoring of the Past project and supported by the Swiss National Science Foundation under grant CR- SII5_173719.

5. CONTACT
Maud Ehrmann (EPFL-DHLAB)
Simon Clematide (UZH)

This dataset includes bibliographic information for 501 papers that were published from 2010-April 2017 (time of search) and use online biodiversity databases for research purposes. Our overarching goal in this study is to determine how research uses of biodiversity data developed during a time of unprecedented growth of online data resources. We also determine uses with the highest number of citations, how online occurrence data are linked to other data types, and if/how data quality is addressed. Specifically, we address the following questions:

1.) What primary biodiversity databases have been cited in published research, and which

databases have been cited most often?

2.) Is the biodiversity research community citing databases appropriately, and are

the cited databases currently accessible online?

3.) What are the most common uses, general taxa addressed, and data linkages, and how

have they changed over time?

4.) What uses have the highest impact, as measured through the mean number of citations

per year?

5.) Are certain uses applied more often for plants/invertebrates/vertebrates?

6.) Are links to specific data types associated more often with particular uses?

7.) How often are major data quality issues addressed?

8.) What data quality issues tend to be addressed for the top uses?

Relevant papers for this analysis include those that use online and openly accessible primary occurrence records, or those that add data to an online database. Google Scholar (GS) provides full-text indexing, which was important to identify data sources that often appear buried in the methods section of a paper. Our search was therefore restricted to GS. All authors discussed and agreed upon representative search terms, which were relatively broad to capture a variety of databases hosting primary occurrence records. The terms included: “species occurrence” database (8,800 results), “natural history collection” database (634 results), herbarium database (16,500 results), “biodiversity database” (3,350 results), “primary biodiversity data” database (483 results), “museum collection” database (4,480 results), “digital accessible information” database (10 results), and “digital accessible knowledge” database (52 results)--note that quotations are used as part of the search terms where specific phrases are needed in whole. We downloaded all records returned by each search (or the first 500 if there were more) into a Zotero reference management database. About one third of the 2500 papers in the final dataset were relevant. Three of the authors with specialized knowledge of the field characterized relevant papers using a standardized tagging protocol based on a series of key topics of interest. We developed a list of potential tags and descriptions for each topic, including: database(s) used, database accessibility, scale of study, region of study, taxa addressed, research use of data, other data types linked to species occurrence data, data quality issues addressed, authors, institutions, and funding sources. Each tagged paper was thoroughly checked by a second tagger.

The final dataset of tagged papers allow us to quantify general areas of research made possible by the expansion of online species occurrence databases, and trends over time. Analyses of this data will be published in a separate quantitative review.

Context

The SEE-AI Project Dataset is a collection of small bowel capsule endoscopy (CE) images obtained using the PillCam™ SB 3 (Medtronic, Minneapolis, MN, USA), which is the subject of the present paper (Small Bowel Capsule Endoscopy Examination with Object Detection Artificial Intelligence Model: The SEE-AI Project; paper is currently in submission). This dataset comprises 18,481 images extracted from 523 small bowel capsule endoscopy videos. We annotated 12,3320 images with 23,033 disease lesions and combined them with 6,161 normal mucosa images. The annotations are provided in YOLO format. While automated or assisted reading techniques for small bowel CE are highly desired, current AI models have not yet been able to accurately identify multiple types of clinically relevant lesions from CE images to the same extent as expert physicians. One major reason for this is the presence of a certain number of images that are difficult to annotate and label, and the lack of adequately constructed data sets. In the aforementioned paper, we tested an object detection model using YOLO v5. The annotations were created by us, and we believe that more effective methods for annotation should be further investigated. We hope that this dataset will be useful for future small bowel CE object detection research."

License

We have presented the dataset of the SEE-AI project at Kaggle (https://www.kaggle.com/), the world’s largest data science online community. Our data are licensed under a Creative Commons Attribution 4.0 International (CC BY 4.0) License. The material is free to copy and redistribute in any medium or format and can be remixed, transformed, and built upon for any purpose if appropriate credit is given.

Credit

More details on this data set can be found in the following paper. Please cite this paper when using this dataset. Yokote, A., Umeno, J., Kawasaki, K., Fujioka, S., Fuyuno, Y., Matsuno, Y., Yoshida, Y., Imazu, N., Miyazono, S., Moriyama, T., Kitazono, T. and Torisu, T. (2024), Small bowel capsule endoscopy examination and open access database with artificial intelligence: The SEE-artificial intelligence project. DEN Open, 4: e258. https://doi.org/10.1002/deo2.258

Content

The main content of The SEE-AI Project Dataset includes image data and annotation data. 18,481 images and annotation data in YOLO format are available. The annotations are written in the txt data whose filenames match the image data. There are also empty txt data for images without annotations.

Acknowledgements

We want to thank Department of Medicine and Clinical Science, Kyushu University, for their cooperation in data collection. We also thank Ultralytics for making YOLO ver5 available. The project name of this dataset was changed due to its name duplication. The previous project name was The AICE project. This was changed on May 14, 2023.

Example usage on google colab with sample model weights

https://colab.research.google.com/drive/1mEE5zXq1U9vC01P-qjxHR2kvxr_3Imz0?usp=sharing

Inspiration

We would be grateful if you could consider setting up better annotation and colleting small intestine CE images. We hope that many more facilities will collect CE images in the future, and datasets will become larger.

Citation metrics are widely used and misused. We have created a publicly available database of top-cited scientists that provides standardized information on citations, h-index, co-authorship adjusted hm-index, citations to papers in different authorship positions and a composite indicator (c-score). Separate data are shown for career-long and, separately, for single recent year impact. Metrics with and without self-citations and ratio of citations to citing papers are given and data on retracted papers (based on Retraction Watch database) as well as citations to/from retracted papers have been added. Scientists are classified into 22 scientific fields and 174 sub-fields according to the standard Science-Metrix classification. Field- and subfield-specific percentiles are also provided for all scientists with at least 5 papers. Career-long data are updated to end-of-2024 and single recent year data pertain to citations received during calendar year 2024. The selection is based on the top 100,000 scientists by c-score (with and without self-citations) or a percentile rank of 2% or above in the sub-field. This version (7) is based on the August 1, 2025 snapshot from Scopus, updated to end of citation year 2024. This work uses Scopus data. Calculations were performed using all Scopus author profiles as of August 1, 2025. If an author is not on the list, it is simply because the composite indicator value was not high enough to appear on the list. It does not mean that the author does not do good work. PLEASE ALSO NOTE THAT THE DATABASE HAS BEEN PUBLISHED IN AN ARCHIVAL FORM AND WILL NOT BE CHANGED. The published version reflects Scopus author profiles at the time of calculation. We thus advise authors to ensure that their Scopus profiles are accurate. REQUESTS FOR CORRECIONS OF THE SCOPUS DATA (INCLUDING CORRECTIONS IN AFFILIATIONS) SHOULD NOT BE SENT TO US. They should be sent directly to Scopus, preferably by use of the Scopus to ORCID feedback wizard (https://orcid.scopusfeedback.com/) so that the correct data can be used in any future annual updates of the citation indicator databases. The c-score focuses on impact (citations) rather than productivity (number of publications) and it also incorporates information on co-authorship and author positions (single, first, last author). If you have additional questions, see attached file on FREQUENTLY ASKED QUESTIONS. Finally, we alert users that all citation metrics have limitations and their use should be tempered and judicious. For more reading, we refer to the Leiden manifesto: https://www.nature.com/articles/520429a

Introduction

This note describes the data sets used for all analyses contained in the manuscript 'Oxytocin - a social peptide?’[1] that is currently under review.

Data Collection

The data sets described here were originally retrieved from Web of Science (WoS) Core Collection via the University of Edinburgh’s library subscription [2]. The aim of the original study for which these data were gathered was to survey peer-reviewed primary studies on oxytocin and social behaviour. To capture relevant papers, we used the following query:

TI = (“oxytocin” OR “pitocin” OR “syntocinon”) AND TS = (“social*” OR “pro$social” OR “anti$social”)

The final search was performed on the 13 September 2021. This returned a total of 2,747 records, of which 2,049 were classified by WoS as ‘articles’. Given our interest in primary studies only – articles reporting original data – we excluded all other document types. We further excluded all articles sub-classified as ‘book chapters’ or as ‘proceeding papers’ in order to limit our analysis to primary studies published in peer-reviewed academic journals. This reduced the set to 1,977 articles. All of these were published in the English language, and no further language refinements were unnecessary.

All available metadata on these 1,977 articles was exported as plain text ‘flat’ format files in four batches, which we later merged together via Notepad++. Upon manually examination, we discovered examples of papers classified as ‘articles’ by WoS that were, in fact, reviews. To further filter our results, we searched all available PMIDs in PubMed (1,903 had associated PMIDs - ~96% of set). We then filtered results to identify all records classified as ‘review’, ‘systematic review’, or ‘meta-analysis’, identifying 75 records 3. After examining a sample and agreeing with the PubMed classification, these were removed these from our dataset - leaving a total of 1,902 articles.

From these data, we constructed two datasets via parsing out relevant reference data via the Sci2 Tool [4]. First, we constructed a ‘node-attribute-list’ by first linking unique reference strings (‘Cite Me As’ column in WoS data files) to unique identifiers, we then parsed into this dataset information on the identify of a paper, including the title of the article, all authors, journal publication, year of publication, total citations as recorded from WoS, and WoS accession number. Second, we constructed an ‘edge-list’ that records the citations from a citing paper in the ‘Source’ column and identifies the cited paper in the ‘Target’ column, using the unique identifies as described previously to link these data to the node-attribute-list.

We then constructed a network in which papers are nodes, and citation links between nodes are directed edges between nodes. We used Gephi Version 0.9.2 [5] to manually clean these data by merging duplicate references that are caused by different reference formats or by referencing errors. To do this, we needed to retain both all retrieved records (1,902) as well as including all of their references to papers whether these were included in our original search or not. In total, this produced a network of 46,633 nodes (unique reference strings) and 112,520 edges (citation links). Thus, the average reference list size of these articles is ~59 references. The mean indegree (within network citations) is 2.4 (median is 1) for the entire network reflecting a great diversity in referencing choices among our 1,902 articles.

After merging duplicates, we then restricted the network to include only articles fully retrieved (1,902), and retrained only those that were connected together by citations links in a large interconnected network (i.e. the largest component). In total, 1,892 (99.5%) of our initial set were connected together via citation links, meaning a total of ten papers were removed from the following analysis – and these were neither connected to the largest component, nor did they form connections with one another (i.e. these were ‘isolates’).

This left us with a network of 1,892 nodes connected together by 26,019 edges. It is this network that is described by the ‘node-attribute-list’ and ‘edge-list’ provided here. This network has a mean in-degree of 13.76 (median in-degree of 4). By restricting our analysis in this way, we lose 44,741 unique references (96%) and 86,501 citations (77%) from the full network, but retain a set of articles tightly knitted together, all of which have been fully retrieved due to possessing certain terms related to oxytocin AND social behaviour in their title, abstract, or associated keywords.

Before moving on, we calculated indegree for all nodes in this network – this counts the number of citations to a given paper from other papers within this network – and have included this in the node-attribute-list. We further clustered this network via modularity maximisation via the Leiden algorithm [6]. We set the algorithm to resolution 1, and allowed the algorithm to run over 100 iterations and 100 restarts. This gave Q=0.43 and identified seven clusters, which we describe in detail within the body of the paper. We have included cluster membership as an attribute in the node-attribute-list.

Data description

We include here two datasets: (i) ‘OTSOC-node-attribute-list.csv’ consists of the attributes of 1,892 primary articles retrieved from WoS that include terms indicating a focus on oxytocin and social behaviour; (ii) ‘OTSOC-edge-list.csv’ records the citations between these papers. Together, these can be imported into a range of different software for network analysis; however, we have formatted these for ease of upload into Gephi 0.9.2. Below, we detail their contents:

1. ‘OTSOC-node-attribute-list.csv’ is a comma-separate values file that contains all node attributes for the citation network (n=1,892) analysed in the paper. The columns refer to:

Id, the unique identifier

Label, the reference string of the paper to which the attributes in this row correspond. This is taken from the ‘Cite Me As’ column from the original WoS download. The reference string is in the following format: last name of first author, publication year, journal, volume, start page, and DOI (if available).

Wos_id, unique Web of Science (WoS) accession number. These can be used to query WoS to find further data on all papers via the ‘UT= ’ field tag.

Title, paper title.

Authors, all named authors.

Journal, journal of publication.

Pub_year, year of publication.

Wos_citations, total number of citations recorded by WoS Core Collection to a given paper as of 13 September 2021

Indegree, the number of within network citations to a given paper, calculated for the network shown in Figure 1 of the manuscript.

Cluster, provides the cluster membership number as discussed within the manuscript (Figure 1). This was established via modularity maximisation via the Leiden algorithm (Res 1; Q=0.43|7 clusters)

1. ‘OTSOC-edge -list.csv’ is a comma-separate values file that contains all citation links between the 1,892 articles (n=26,019). The columns refer to:

Source, the unique identifier of the citing paper.

Target, the unique identifier of the cited paper.

Type, edges are ‘Directed’, and this column tells Gephi to regard all edges as such.

Syr_date, this contains the date of publication of the citing paper.

Tyr_date, this contains the date of publication of the cited paper.

Software recommended for analysis

Gephi version 0.9.2 was used for the visualisations within the manuscript, and both files can be read and into Gephi without modification.

Notes

[1] Leng, G., Leng, R. I., Ludwig, M. (Submitted). Oxytocin – a social peptide? Deconstructing the evidence.

[2] Edinburgh University’s subscription to Web of Science covers the following databases: (i) Science Citation Index Expanded, 1900-present; (ii) Social Sciences Citation Index, 1900-present; (iii) Arts & Humanities Citation Index, 1975-present; (iv) Conference Proceedings Citation Index- Science, 1990-present; (v) Conference Proceedings Citation Index- Social Science & Humanities, 1990-present; (vi) Book Citation Index– Science, 2005-present; (vii) Book Citation Index– Social Sciences & Humanities, 2005-present; (viii) Emerging Sources Citation Index, 2015-present.

[3] For those interested, the following PMIDs were identified as ‘articles’ by WoS, but as ‘reviews’ by PubMed: ‘34502097’ ‘33400920’ ‘32060678’ ‘31925983’ ‘31734142’ ‘30496762’ ‘30253045’ ‘29660735’ ‘29518698’ ‘29065361’ ‘29048602’ ‘28867943’ ‘28586471’ ‘28301323’ ‘27974283’ ‘27626613’ ‘27603523’ ‘27603327’ ‘27513442’ ‘27273834’ ‘27071789’ ‘26940141’ ‘26932552’ ‘26895254’ ‘26869847’ ‘26788924’ ‘26581735’ ‘26548910’ ‘26317636’ ‘26121678’ ‘26094200’ ‘25997760’ ‘25631363’ ‘25526824’ ‘25446893’ ‘25153535’ ‘25092245’ ‘25086828’ ‘24946432’ ‘24637261’ ‘24588761’ ‘24508579’ ‘24486356’ ‘24462936’ ‘24239932’ ‘24239931’ ‘24231551’ ‘24216134’ ‘23955310’ ‘23856187’ ‘23686025’ ‘23589638’ ‘23575742’ ‘23469841’ ‘23055480’ ‘22981649’ ‘22406388’ ‘22373652’ ‘22141469’ ‘21960250’ ‘21881219’ ‘21802859’ ‘21714746’ ‘21618004’ ‘21150165’ ‘20435805’ ‘20173685’ ‘19840865’ ‘19546570’ ‘19309413’ ‘15288368’ ‘12359512’ ‘9401603’ ‘9213136’ ‘7630585’

[4] Sci2 Team. (2009). Science of Science (Sci2) Tool. Indiana University and SciTech Strategies. Stable URL: https://sci2.cns.iu.edu

[5] Bastian, M., Heymann, S., & Jacomy, M. (2009).

This dataset contains data collected during a study "Understanding the development of public data ecosystems: from a conceptual model to a six-generation model of the evolution of public data ecosystems" conducted by Martin Lnenicka (University of Hradec Králové, Czech Republic), Anastasija Nikiforova (University of Tartu, Estonia), Mariusz Luterek (University of Warsaw, Warsaw, Poland), Petar Milic (University of Pristina - Kosovska Mitrovica, Serbia), Daniel Rudmark (Swedish National Road and Transport Research Institute, Sweden), Sebastian Neumaier (St. Pölten University of Applied Sciences, Austria), Karlo Kević (University of Zagreb, Croatia), Anneke Zuiderwijk (Delft University of Technology, Delft, the Netherlands), Manuel Pedro Rodríguez Bolívar (University of Granada, Granada, Spain).

As there is a lack of understanding of the elements that constitute different types of value-adding public data ecosystems and how these elements form and shape the development of these ecosystems over time, which can lead to misguided efforts to develop future public data ecosystems, the aim of the study is: (1) to explore how public data ecosystems have developed over time and (2) to identify the value-adding elements and formative characteristics of public data ecosystems. Using an exploratory retrospective analysis and a deductive approach, we systematically review 148 studies published between 1994 and 2023. Based on the results, this study presents a typology of public data ecosystems and develops a conceptual model of elements and formative characteristics that contribute most to value-adding public data ecosystems, and develops a conceptual model of the evolutionary generation of public data ecosystems represented by six generations called Evolutionary Model of Public Data Ecosystems (EMPDE). Finally, three avenues for a future research agenda are proposed.

This dataset is being made public both to act as supplementary data for "Understanding the development of public data ecosystems: from a conceptual model to a six-generation model of the evolution of public data ecosystems ", Telematics and Informatics*, and its Systematic Literature Review component that informs the study.

Description of the data in this data set

PublicDataEcosystem_SLR provides the structure of the protocol

Spreadsheet#1 provides the list of results after the search over three indexing databases and filtering out irrelevant studies

Spreadsheets #2 provides the protocol structure.

Spreadsheets #3 provides the filled protocol for relevant studies.

The information on each selected study was collected in four categories:(1) descriptive information,(2) approach- and research design- related information,(3) quality-related information,(4) HVD determination-related information

Descriptive Information

Article number

A study number, corresponding to the study number assigned in an Excel worksheet

Complete reference

The complete source information to refer to the study (in APA style), including the author(s) of the study, the year in which it was published, the study's title and other source information.

Year of publication

The year in which the study was published.

Journal article / conference paper / book chapter

The type of the paper, i.e., journal article, conference paper, or book chapter.

Journal / conference / book

Journal article, conference, where the paper is published.

DOI / Website

A link to the website where the study can be found.

Number of words

A number of words of the study.

Number of citations in Scopus and WoS

The number of citations of the paper in Scopus and WoS digital libraries.

Availability in Open Access

Availability of a study in the Open Access or Free / Full Access.

Keywords

Keywords of the paper as indicated by the authors (in the paper).

Relevance for our study (high / medium / low)

What is the relevance level of the paper for our study

Approach- and research design-related information

Approach- and research design-related information

Objective / Aim / Goal / Purpose & Research Questions

The research objective and established RQs.

Research method (including unit of analysis)

The methods used to collect data in the study, including the unit of analysis that refers to the country, organisation, or other specific unit that has been analysed such as the number of use-cases or policy documents, number and scope of the SLR etc.

Study’s contributions

The study’s contribution as defined by the authors

Qualitative / quantitative / mixed method

Whether the study uses a qualitative, quantitative, or mixed methods approach?

Availability of the underlying research data

Whether the paper has a reference to the public availability of the underlying research data e.g., transcriptions of interviews, collected data etc., or explains why these data are not openly shared?

Period under investigation

Period (or moment) in which the study was conducted (e.g., January 2021-March 2022)

Use of theory / theoretical concepts / approaches? If yes, specify them

Does the study mention any theory / theoretical concepts / approaches? If yes, what theory / concepts / approaches? If any theory is mentioned, how is theory used in the study? (e.g., mentioned to explain a certain phenomenon, used as a framework for analysis, tested theory, theory mentioned in the future research section).

Quality-related information

Quality concerns

Whether there are any quality concerns (e.g., limited information about the research methods used)?

Public Data Ecosystem-related information

Public data ecosystem definition

How is the public data ecosystem defined in the paper and any other equivalent term, mostly infrastructure. If an alternative term is used, how is the public data ecosystem called in the paper?

Public data ecosystem evolution / development

Does the paper define the evolution of the public data ecosystem? If yes, how is it defined and what factors affect it?

What constitutes a public data ecosystem?

What constitutes a public data ecosystem (components & relationships) - their "FORM / OUTPUT" presented in the paper (general description with more detailed answers to further additional questions).

Components and relationships

What components does the public data ecosystem consist of and what are the relationships between these components? Alternative names for components - element, construct, concept, item, helix, dimension etc. (detailed description).

Stakeholders

What stakeholders (e.g., governments, citizens, businesses, Non-Governmental Organisations (NGOs) etc.) does the public data ecosystem involve?

Actors and their roles

What actors does the public data ecosystem involve? What are their roles?

Data (data types, data dynamism, data categories etc.)

What data do the public data ecosystem cover (is intended / designed for)? Refer to all data-related aspects, including but not limited to data types, data dynamism (static data, dynamic, real-time data, stream), prevailing data categories / domains / topics etc.

Processes / activities / dimensions, data lifecycle phases

What processes, activities, dimensions and data lifecycle phases (e.g., locate, acquire, download, reuse, transform, etc.) does the public data ecosystem involve or refer to?

Level (if relevant)

What is the level of the public data ecosystem covered in the paper? (e.g., city, municipal, regional, national (=country), supranational, international).

Other elements or relationships (if any)

What other elements or relationships does the public data ecosystem consist of?

Additional comments

Additional comments (e.g., what other topics affected the public data ecosystems and their elements, what is expected to affect the public data ecosystems in the future, what were important topics by which the period was characterised etc.).

New papers

Does the study refer to any other potentially relevant papers?

Additional references to potentially relevant papers that were found in the analysed paper (snowballing).

Format of the file.xls, .csv (for the first spreadsheet only), .docx

Licenses or restrictionsCC-BY

For more info, see README.txt

Conceptual novelty analysis data based on PubMed Medical Subject Headings ---------------------------------------------------------------------- Created by Shubhanshu Mishra, and Vetle I. Torvik on April 16th, 2018 ## Introduction This is a dataset created as part of the publication titled: Mishra S, Torvik VI. Quantifying Conceptual Novelty in the Biomedical Literature. D-Lib magazine : the magazine of the Digital Library Forum. 2016;22(9-10):10.1045/september2016-mishra. It contains final data generated as part of our experiments based on MEDLINE 2015 baseline and MeSH tree from 2015. The dataset is distributed in the form of the following tab separated text files: * PubMed2015_NoveltyData.tsv - Novelty scores for each paper in PubMed. The file contains 22,349,417 rows and 6 columns, as follow: - PMID: PubMed ID - Year: year of publication - TimeNovelty: time novelty score of the paper based on individual concepts (see paper) - VolumeNovelty: volume novelty score of the paper based on individual concepts (see paper) - PairTimeNovelty: time novelty score of the paper based on pair of concepts (see paper) - PairVolumeNovelty: volume novelty score of the paper based on pair of concepts (see paper) * mesh_scores.tsv - Temporal profiles for each MeSH term for all years. The file contains 1,102,831 rows and 5 columns, as follow: - MeshTerm: Name of the MeSH term - Year: year - AbsVal: Total publications with that MeSH term in the given year - TimeNovelty: age (in years since first publication) of MeSH term in the given year - VolumeNovelty: : age (in number of papers since first publication) of MeSH term in the given year * meshpair_scores.txt.gz (36 GB uncompressed) - Temporal profiles for each MeSH term for all years - Mesh1: Name of the first MeSH term (alphabetically sorted) - Mesh2: Name of the second MeSH term (alphabetically sorted) - Year: year - AbsVal: Total publications with that MeSH pair in the given year - TimeNovelty: age (in years since first publication) of MeSH pair in the given year - VolumeNovelty: : age (in number of papers since first publication) of MeSH pair in the given year * README.txt file ## Dataset creation This dataset was constructed using multiple datasets described in the following locations: * MEDLINE 2015 baseline: https://www.nlm.nih.gov/bsd/licensee/2015_stats/baseline_doc.html * MeSH tree 2015: ftp://nlmpubs.nlm.nih.gov/online/mesh/2015/meshtrees/ * Source code provided at: https://github.com/napsternxg/Novelty Note: The dataset is based on a snapshot of PubMed (which includes Medline and PubMed-not-Medline records) taken in the first week of October, 2016. Check here for information to get PubMed/MEDLINE, and NLMs data Terms and Conditions: Additional data related updates can be found at: Torvik Research Group ## Acknowledgments This work was made possible in part with funding to VIT from NIH grant P01AG039347 and NSF grant 1348742 . The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. ## License Conceptual novelty analysis data based on PubMed Medical Subject Headings by Shubhanshu Mishra, and Vetle Torvik is licensed under a Creative Commons Attribution 4.0 International License. Permissions beyond the scope of this license may be available at https://github.com/napsternxg/Novelty

Background: The assessment of scientific publications is an integral part of the scientific process. Here we investigate three methods of assessing the merit of a scientific paper: subjective post-publication peer review, the number of citations gained by a paper and the impact factor of the journal in which the article was published. Methodology/principle findings: We investigate these methods using two datasets in which subjective post-publication assessments of scientific publications have been made by experts. We find that there are moderate, but statistically significant, correlations between assessor scores, when two assessors have rated the same paper, and between assessor score and the number of citations a paper accrues. However, we show that assessor score depends strongly on the journal in which the paper is published, and that assessors tend to over-rate papers published in journals with high impact factors. If we control for this bias, we find that the correlation between assessor scores and between assessor score and the number of citations is weak, suggesting that scientists have little ability to judge either the intrinsic merit of a paper or its likely impact. We also show that the number of citations a paper receives is an extremely error-prone measure of scientific merit. Finally, we argue that the impact factor is likely to be a poor measure of merit, since it depends on subjective assessment. Conclusions: We conclude that the three measures of scientific merit considered here are poor; in particular subjective assessments are an error-prone, biased and expensive method by which to assess merit. We argue that the impact factor may be the most satisfactory of the methods we have considered, since it is a form of pre-publication review. However, we emphasise that it is likely to be a very error-prone measure of merit that is qualitative, not quantitative.

Abstract:

In recent years there has been an increased interest in Artificial Intelligence for IT Operations (AIOps). This field utilizes monitoring data from IT systems, big data platforms, and machine learning to automate various operations and maintenance (O&M) tasks for distributed systems.
The major contributions have been materialized in the form of novel algorithms.
Typically, researchers took the challenge of exploring one specific type of observability data sources, such as application logs, metrics, and distributed traces, to create new algorithms.
Nonetheless, due to the low signal-to-noise ratio of monitoring data, there is a consensus that only the analysis of multi-source monitoring data will enable the development of useful algorithms that have better performance.
Unfortunately, existing datasets usually contain only a single source of data, often logs or metrics. This limits the possibilities for greater advances in AIOps research.
Thus, we generated high-quality multi-source data composed of distributed traces, application logs, and metrics from a complex distributed system. This paper provides detailed descriptions of the experiment, statistics of the data, and identifies how such data can be analyzed to support O&M tasks such as anomaly detection, root cause analysis, and remediation.

General Information:

This repository contains the simple scripts for data statistics, and link to the multi-source distributed system dataset.

You may find details of this dataset from the original paper:

Sasho Nedelkoski, Jasmin Bogatinovski, Ajay Kumar Mandapati, Soeren Becker, Jorge Cardoso, Odej Kao, "Multi-Source Distributed System Data for AI-powered Analytics".

If you use the data, implementation, or any details of the paper, please cite!

BIBTEX:

_

@inproceedings{nedelkoski2020multi,
 title={Multi-source Distributed System Data for AI-Powered Analytics},
 author={Nedelkoski, Sasho and Bogatinovski, Jasmin and Mandapati, Ajay Kumar and Becker, Soeren and Cardoso, Jorge and Kao, Odej},
 booktitle={European Conference on Service-Oriented and Cloud Computing},
 pages={161--176},
 year={2020},
 organization={Springer}
}

_

The multi-source/multimodal dataset is composed of distributed traces, application logs, and metrics produced from running a complex distributed system (Openstack). In addition, we also provide the workload and fault scripts together with the Rally report which can serve as ground truth. We provide two datasets, which differ on how the workload is executed. The sequential_data is generated via executing workload of sequential user requests. The concurrent_data is generated via executing workload of concurrent user requests.

The raw logs in both datasets contain the same files. If the user wants the logs filetered by time with respect to the two datasets, should refer to the timestamps at the metrics (they provide the time window). In addition, we suggest to use the provided aggregated time ranged logs for both datasets in CSV format.

Important: The logs and the metrics are synchronized with respect time and they are both recorded on CEST (central european standard time). The traces are on UTC (Coordinated Universal Time -2 hours). They should be synchronized if the user develops multimodal methods. Please read the IMPORTANT_experiment_start_end.txt file before working with the data.

Our GitHub repository with the code for the workloads and scripts for basic analysis can be found at: https://github.com/SashoNedelkoski/multi-source-observability-dataset/

Neural Information Processing Systems (NIPS) is one of the top machine learning conferences in the world. It covers topics ranging from deep learning and computer vision to cognitive science and reinforcement learning.

This year, Kaggle is hosting the NIPS 2015 paper dataset to facilitate and showcase exploratory analytics on the NIPS data. We've extracted the paper text from the raw PDF files and are releasing that both in CSV files and as a SQLite database. Here's a quick script that gives an overview of what's included in the data.

We encourage you to explore this data and share what you find through Kaggle Scripts!

Data Description

Overview of the data in Kaggle Scripts.

nips-2015-papers-release-*.zip (downloadable from the link above) contains the below files/folders. All this data's available through Kaggle Scripts as well, and you can create a new script to immediately start exploring the data in R, Python, Julia, or SQLite.

This dataset is available in two formats: three CSV files and a single SQLite database (consisting of three tables with content identical to the CSV files).

You can see the code used to create this dataset on Github.

Papers.csv

This file contains one row for each of the 403 NIPS papers from this year's conference. It includes the following fields

• Id - unique identifier for the paper (equivalent to the one in NIPS's system)
• Title - title of the paper
• EventType - whether it's a poster, oral, or spotlight presentation
• PdfName - filename for the PDF document
• Abstract - text for the abstract (scraped from the NIPS website)
• PaperText - raw text from the PDF document (created using the tool pdftotext)

Authors.csv

This file contains id's and names for each of the authors on this year's NIPS papers.

• Id - unique identifier for the author (equivalent to the one in NIPS's system)
• Name - author's name

PaperAuthors.csv

This file links papers to their corresponding authors.

• Id - unique identifier
• PaperId - id for the paper
• AuthorId - id for the author

database.sqlite

This SQLite database contains the tables with equivalent data and formatting as the Papers.csv, Authors.csv, and PaperAuthors.csv files.

pdfs

This folder contains the raw pdf files for each of the papers.

Abstract (en): This repository contains the data and code necessary to replicate all figures and tables in the working paper: "Does the disclosure of gun ownership affect crime? Evidence from New York" by Daniel TannenbaumThere are four folders in this repository:(1) Build: contains all the .do files required to produce the analysis datasets, using the raw data (i.e. datasets in the RawData folder).(2) Analysis: contains all the .do files required to produce all the figures and tables in the paper, using the analysis datasets (i.e. datasets in the AnalysisData folder).(3) RawData: contains all the raw datasets used to produce the AnalysisData datasets. The only raw dataset used in the paper that is excluded from this folder is the proprietary housing assessor and sales transaction data from DataQuick, owned by Corelogic. If I receive approval to include this raw data in this repository I will do so in future versions of this repository.(4) AnalysisData: contains all the analysis datasets that are created using the Build and are used to produce the tables and figures in the paper.Running the file Master_analysis.do in the Analysis folder will produce, in one script, all the tables and figures in the paper.

Overview

The OpenResume dataset is designed for researchers and practitioners in career trajectory modeling and job-domain machine learning, as described in the IEEE BigData 2024 paper. It includes both anonymized realistic resumes and synthetically generated resumes, offering a comprehensive resource for developing and benchmarking predictive models across a variety of career-related tasks. By employing anonymization and differential privacy techniques, OpenResume ensures that research can be conducted while maintaining privacy. The dataset is available in this repository. Please see the paper for more details: 10.1109/BigData62323.2024.10825519

If you find this paper useful in your research or use this dataset in any publications, projects, tools, or other forms, please cite:

@inproceedings{yamashita2024openresume,

title={{OpenResume: Advancing Career Trajectory Modeling with Anonymized and Synthetic Resume Datasets}},

author={Yamashita, Michiharu and Tran, Thanh and Lee, Dongwon},

booktitle={2024 IEEE International Conference on Big Data (BigData)},

year={2024},

organization={IEEE}

}

@inproceedings{yamashita2023james,

title={{JAMES: Normalizing Job Titles with Multi-Aspect Graph Embeddings and Reasoning}},

author={Yamashita, Michiharu and Shen, Jia Tracy and Tran, Thanh and Ekhtiari, Hamoon and Lee, Dongwon},

booktitle={2023 IEEE International Conference on Data Science and Advanced Analytics (DSAA)},

year={2023},

organization={IEEE}

}

LitSearch: A Retrieval Benchmark for Scientific Literature Search

This dataset contains the query set and retrieval corpus for our paper LitSearch: A Retrieval Benchmark for Scientific Literature Search. We introduce LitSearch, a retrieval benchmark comprising 597 realistic literature search queries about recent ML and NLP papers. LitSearch is constructed using a combination of (1) questions generated by GPT-4 based on paragraphs containing inline citations from research papers and… See the full description on the dataset page: https://huggingface.co/datasets/princeton-nlp/LitSearch.

Abstract: These are the experimental data for the paper Bach, Jakob. "Using Constraints to Discover Sparse and Alternative Subgroup Descriptions" published on arXiv in 2024. You can find the paper here and the code here. See the README for details. The datasets used in our study (which we also provide here) originate from PMLB. The corresponding GitHub repository is MIT-licensed ((c) 2016 Epistasis Lab at UPenn). Please see the file LICENSE in the folder datasets/ for the license text. TechnicalRemarks: # Experimental Data for the Paper "Using Constraints to Discover Sparse and Alternative Subgroup Descriptions" These are the experimental data for the paper Bach, Jakob. "Using Constraints to Discover Sparse and Alternative Subgroup Descriptions"

Multivariate Time-Series (MTS) are ubiquitous, and are generated in areas as disparate as sensor recordings in aerospace systems, music and video streams, medical monitoring, and financial systems. Domain experts are often interested in searching for interesting multivariate patterns from these MTS databases which can contain up to several gigabytes of data. Surprisingly, research on MTS search is very limited. Most existing work only supports queries with the same length of data, or queries on a fixed set of variables. In this paper, we propose an efficient and flexible subsequence search framework for massive MTS databases, that, for the first time, enables querying on any subset of variables with arbitrary time delays between them. We propose two provably correct algorithms to solve this problem — (1) an R-tree Based Search (RBS) which uses Minimum Bounding Rectangles (MBR) to organize the subsequences, and (2) a List Based Search (LBS) algorithm which uses sorted lists for indexing. We demonstrate the performance of these algorithms using two large MTS databases from the aviation domain, each containing several millions of observations. Both these tests show that our algorithms have very high prune rates (>95%) thus needing actual disk access for only less than 5% of the observations. To the best of our knowledge, this is the first flexible MTS search algorithm capable of subsequence search on any subset of variables. Moreover, MTS subsequence search has never been attempted on datasets of the size we have used in this paper.