Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
18
datasets available to search
ShareScore release 0.9.0
Dataset results
18 results for “metadata model”
SARS-CoV-2 wastewater surveillance data and metadata in the Open Data Model format. Part 1: Québec City
<p>SARS-CoV-2 wastewater surveillance data and metadata in the Open Data Model format. Part 1: Québec City Authors</p> <ul> <li>Therrien, J-D<sup>1</sup></li> <li>Maere, T.<sup>1</sup></li> <li>Sanchez-Quete, F.<sup>2</sup></li> <li>Tsitouras, A.<sup>2</sup></li> <li>Goitom, E.<sup>3</sup></li> <li>Cloutier, F.<sup>4</sup></li> <li>Dufour, D.<sup>4</sup></li> <li>Proulx, F. <sup>4</sup></li> <li>Nicolaï, N.<sup>1</sup></li> <li>Philippe, R.<sup>1</sup></li> <li>Tohidi, M.<sup>1</sup></li> <li>Dorner, S.<sup>3</sup></li> <li>Frigon, D.<sup>2</sup></li> <li>Vanrolleghem, P.A.<sup>1</sup></li> </ul> <p>Affiliations</p> <ul> <li><sup>1</sup> model<em>EAU</em>, Département de génie civil et de génie des eaux, Université Laval</li> <li><sup>2</sup> Microbial Community Engineering Lab (MiCEL), Department of Civil Engineering, McGill University</li> <li><sup>3</sup> Polytechnique Montréal</li> <li><sup>4</sup> Ville de Québec</li> </ul> <p>General Remarks</p> <p>Wastewater-based surveillance of SARS-CoV-2 virus can detect between 1 and 30 infected individuals per 100,000 (including asymptomatic ones) by analyzing the population's sewage. As such, this method is very attractive since it costs only a fraction of clinical testing (as low as 1%). Human faeces may contain the virus a few days before a person becomes ill. Thus, this approach allows for detection of outbreaks 2-7 days before the increase in reported cases stemming from clinical screening tests (Bibby et al., 2021). Wastewater-based surveillance complements clinical testing by geolocating outbreaks, which may help targeting intensive screening programs. Moreover, it provides a quick indication of whether new public health measures (e.g., masks, social distancing, confinement, and curfew) are effective.</p> <p>Sampling</p> <p>The reported dataset contains open data collected in the province of Québec as part of the SARS-CoV-2 wastewater-based surveillance program <a href="https://www.centreau.ulaval.ca/en/covid/">CentrEau</a>-COVID. Four of the largest cities in the province (Montréal, Laval, Québec City, and Trois-Rivières), as well as the municipalities of four rural regions (Mauricie, Centre-du-Québec, Bas-St-Laurent, and Gaspésie) participated in the program. The entire dataset includes 31 sampling sites covering approximately half the population of the province of Québec (population size of 8.5 million). The timeframe covered by the dataset varies for each site. The earliest surveillance program was launched in March 2020, others followed soon after. Samples were collected using various methods, such as 24h composite samples, grab samples, and passive sampling using variations on the Moore swab method (Schang et al., 2020)</p> <p>Analysis</p> <p>Prior to the analysis of the samples for SARS-CoV-2, physiochemical parameters such as total suspended solids (TSS), turbidity, conductivity, ammonium concentration, and pH were measured. The samples were subsequently concentred by filtration using a MEC filter (0.45 um), followed by total RNA extraction using the Qiagen AllPrep PowerViral DNA/RNA Kit (Qiagen, USA) with some modifications (beta-mercaptoethanol concentration raised to 10% and lysis performed at 55 °C for 30 minutes) (Ahmed et al., 2020). SARS-CoV-2 viral RNA was detected by a one-step RT-qPCR. To assess the RNA recovery rate of the procedure, samples were spiked before extraction with a known concentration of Bovine Respiratory Syncytial Virus (BRSV) using the Zoetis INFORCE 3 vaccine (Zoetis, USA). In addition to SARS-CoV-2, samples were assessed for Pepper Mild Mottle Virus (PMMoV), the daily load of which is hypothesized to represent the fecal load contributions to the samples at a given site and time. PCR conditions and primer used to collect viral data are described in the files <code>primers.md</code> and <code>PCR conditions.md</code>.</p> <p>Compilation</p> <p>The measurements on wastewater samples carried out by the participating laboratories of this study are found in the <code>WWMeasure</code> table. The values provided by municipalities come from laboratories accredited by the Centre d'expertise en analyse environnementale du Québec (CEAEQ), in compliance with the latter's quality assurance protocols. The COVID-19-related public health data found in the <code>CPHD</code> table were collected from the Institut National de Santé Publique du Québec (INSPQ)'s public reports. Wastewater data taken in-situ at the sampling sites (e.g., the flow at pumping stations or water resource recovery facilities (WRRFs)) are found in the <code>SiteMeasure</code> table and were taken by the institutions responsible for managing the sites. All of the data, stemming from multiple sources, were combined into the <a href="https://github.com/Big-Life-Lab/ODM">Open Data Model (ODM)</a> standard format using the <a href="https://github.com/modelEAU/ODM-Import">ODM-Import python package</a> (see also Structure).</p> <p>Validation</p> <p>Wastewater and sample data were manually assessed for quality by our research collaborators. Data points for which the quality appeared to be uncertain were tagged with the value <code>True</code> in the <code>qualityFlag</code> column. Conversely, data deemed of good quality have a quality flag of <code>False</code>. Data that were not checked have a quality flag of <code>NA</code>. Textual comments describing the issues with the data points in more detail are also included in the dataset using the <code>notes</code> column of the relevant tables. Note that data validation was carried out by the data custodians responsible for each city in the dataset according to available resources. As the project continues and data validation is undertaken on more sections of the dataset, data may be re-analyzed, flagged, or commented as needed. Revisions to the dataset will be reported to the best of our ability.</p> <p>Structure</p> <p>The data contained in this dataset has been structured according to the <a href="https://github.com/Big-Life-Lab/ODM">Open Data Model (ODM) for Wastewater-Based Surveillance</a>. This model provides a standardized dictionary to collect and share data and metadata stemming from wastewater-based surveillance programs. By convention, it splits all data into 10+ thematic tables with each record representing a unique measurement, i.e., long format. For convenience, the <code>wide</code> folder presents the data found in all the other tables in a wide format, i.e., multiple measurements are aligned by <code>timestamp</code>, with each column representing a different parameter.</p> <p>Acknowledgements</p> <p>The authors would like to acknowledge that this dataset was collected thanks to the financial support of the Fonds de Recherche du Québec, the Molson Foundation, the Trottier Family Foundation, CentrEau and NSERC. The authors would also like to acknowledge the efforts of Douglas Manuel (Ottawa Hospital) and Howard Swerdfeger (Public Health Agency of Canada) for their original idea for the Open Data Model and continued development.</p> <p>References</p> <ol> <li> <p>Ahmed, W., Bertsch, P.M., Bivins, A., Bibby, K., Farkas, K., Gathercole, A., Haramoto, E., Gyawali, P., Korajkic, A., McMinn, B.R., Mueller, J.F., Simpson, S.L., Smith, W.J.M., Symonds, E.M., Thomas, K. v., Verhagen, R., Kitajima, M., 2020. Comparison of virus concentration methods for the RT-qPCR-based recovery of murine hepatitis virus, a surrogate for SARS-CoV-2 from untreated wastewater. Science of the Total Environment 739. <a href="https://doi.org/10.1016/j.scitotenv.2020.139960">https://doi.org/10.1016/j.scitotenv.2020.139960</a></p> </li> <li> <p>Bibby, K., Bivins, A., Wu, Z., North, D., 2021. Making waves: Plausible lead time for wastewater based epidemiology as an early warning system for COVID-19. Water Research 202, 117438. <a href="https://doi.org/10.1016/j.watres.2021.117438">https://doi.org/10.1016/j.watres.2021.117438</a></p> </li> <li> <p>Schang, C., Crosbie, N., Nolan, M., Poon, R., Wang, M., Jex, A., Scales, P., Schmidt, J., Thorley, B.R., Henry, R., Kolotelo, P., Langeveld, J., Schilperoort, R., Shi, B., Einsiedel, S., Thomas, M., Black, J., Wilson, S., McCarthy, D.T., 2020. Passive sampling of viruses for wastewater-based epidemiology: a case-study of SARS-CoV-2 [WWW Document]. URL <a href="https://www.researchgate.net/publication/347103410\_Passive\_sampling\_of\_viruses\_for\_wastewater-based\_epidemiology\_a\_case-study\_of\_SARS-CoV-2?channel=doi&linkId=5fd800f392851c13fe892393&showFulltext=true">https://www.researchgate.net/publication/347103410\_Passive\_sampling\_of\_viruses\_for\_wastewater-based\_epidemiology\_a\_case-study\_of\_SARS-CoV-2?channel=doi&linkId=5fd800f392851c13fe892393&showFulltext=true</a> (accessed 1.18.21).</p> </li> </ol>
Metadata dataset: benchmark datasets for modelling
<p>The main goal of the Soil Mission MARVIC project is to develop a framework for designing harmonized context-specific Monitoring, Reporting and Verification (MRV) systems for carbon farming, in support of the EU Carbon Removals and Carbon Farming (CRCF) regulation.</p> <p>The scope of this report (MARVIC Deliverable 2.1) is to provide a metadata dataset of benchmark sites (BS) that are relevant to the calibration and validation of models used within the MARVIC test cases. The dataset provided by Deliverable 2.1 describes the main characteristics of each site, such as pedoclimatic conditions, management practices applied, soil chemical, physical, and biological parameters, and details the measured variables that have been collected over time. </p> <p>This dataset of metadata is used within MARVIC to determine which modelling approaches can be used in each of the test cases across work packages. </p>
AIMMX: Artificial Intelligence Model Metadata Extractor
<p>Replication set that contains evaluation dataset and samples for MSR2020 submitted paper: "AIMMX: Artificial Intelligence Model Metadata Extractor."</p>
Metadata - A New Application for Cenicriviroc, a Dual CCR2/CCR5 Antagonist, in the Treatment of Painful Diabetic Neuropathy in a Mouse Model
Open the record for dataset details and reuse information.
Application Case 3: Data sets consisting of 3D scans, 3D model and the derived metadata, from two different software programs
<p>In this repository we provide 3D scan projects and the 3D models processed from them with their metadata using the example of a wood sample. The metadata was generated using our metadata generation script, which is described in the referenced publication.</p> <p>The 3D scan projects were created in different software (atos v6.2, atos 2016 and zeiss 2023). For each there is a scan project, a 3D model and the generated metadata with and without uri in this repository.</p> <p>The publication in which this application case is included: Homburg, T., Cramer, A., Raddatz, L. <em>et al.</em> Metadata schema and ontology for capturing and processing of 3D cultural heritage objects. <em>Herit Sci</em> <strong>9</strong>, 91 (2021). <a href="https://doi.org/10.1186/s40494-021-00561-w">https://doi.org/10.1186/s40494-021-00561-w</a></p> <p>Python scripts for exporting metadata can be found here: <a href="https://github.com/i3mainz/3dcap-md-gen/tree/0.1.3">GitHub - i3mainz/3dcap-md-gen</a></p>
Case study on an integrated interoperable metadata model for geoscience information resources
<p>An integrated metadata schema was created to promote interoperability and to characterize historically collected geoscience data. The metadata standard describes the steps taken to develop a switching‐across methodology that allows geoscience metadata standards (ISO 19115, CSDGM, ANZLIC and INSPIRE) to communicate with one another. We also demonstrated a real example of one of the switching‐across that was created. A representative metadata schema was created using items from harmonization standards that might be optimally mapped. Semantic interoperability allows users to access equivalent classes across several metadata standards in a uniform and consistent rule. It also tries to provide fundamental aspects that might characterize acquisition information to enhance data reuse and current metadata standards for data preservation and distribution.</p> <p>This dataset contains 5 data and contains the content standard of each of the four metadata schema and the metadata information of integrated metadata schema.</p> <ul> <li>data_ISO19115.xlsx: Core level of metadata elements for geographic resources of ISO 19115</li> <li>data_CSDGM.xlsx: Content Standard for Digital Geospatial Metadata of CSDGM</li> <li>data_INSPIRE.xlsx: Content Standard of INSPIRE metadata modeldata_ANZLIC.xlsx: Content Standard of ANZLIC Metadata Profile</li> <li>data_integrated model.xlsx: metadata class and their elements in the integrated interoperable metadata model</li> </ul>
Modelling and Automated Retrieval of Provenance Relationships (Metadata and Statistics)
<p>This dataset contains the data used in the master's thesis with the above title. It consists of a BibTeX file with the bibliographic metadata of the publications and websites cited throughout the thesis, and a Markdown file with statistics of the data sources discussed in Chapter 4.</p>
Extracting Enhanced Artifical Intelligence Model Metadata from Software Repositories
<p>Replication set for evaluations performed in EMSE submission "Extracting Enhanced Artifical Intelligence Model Metadata from Software Repositories" </p>
Automated Extraction of Artificial Intelligence Model Metadata from Repositories
<p>Replication datasets for MSR submission: "Automated Extraction of Artificial Intelligence Model Metadata from Repositories"</p> <p>This package contains the following:</p> <p>1. The model zoo dataset of 277 repository URLs.<br> 2. The arXiv dataset of 1,398 repository URLs extracted from arXiv papers.<br> 3. The annotated evaluation samples for both datasets with incorrect and missing properties marked per model.<br> </p>
Quantifying niche similarity among new world seed plants--Species Distribution Models (SDMs) & associated metadata
<p>Niche shift and conservatism are often framed as mutually exclusive. However, both processes could contribute to biodiversity patterns. We tested this expectation by quantifying the degree of climatic niche similarity among New World seed plants.</p> <p>To incorporate the biological reality that species experience varied abiotic conditions across their range, we assembled distribution models and used these to characterize temperature, precipitation, and elevation niches for species as continuously-valued distributions. We then quantified niche similarity (distributional overlap) and identified statistically significant differences compared to a randomized null.</p> <p>The degree of niche similarity differed among climate variables, plant lineages, and at different phylogenetic scales. For example, ~17% of all seed plants were significantly different in elevational niche from their closest relative(s), whereas for precipitation, this value was only ~4%. Average niche similarity decreased with increasing phylogenetic distance, consistent with niche conservatism; however, variance in niche similarity among close relatives was large, such that there always existed niche differences equaling those among distantly related species.</p> <p>Our results suggest researchers should incorporate both niche shift and conservatism as important, scale-dependent factors shaping biodiversity patterns as these processes are not mutually exclusive, nor do they contribute equally to patterns among different plant lineages or niche variables.</p>
Quantifying niche similarity among new world seed plants--Species Distribution Models (SDMs) & associated metadata
Open the record for dataset details and reuse information.
Modeling suite prototype and sourcebook Metadata
Open the record for dataset details and reuse information.
Nest survival, female survival, and average covariates used in modeling, along with metadata
<p class="CxSpFirst">Females must balance physiological and behavioral demands of producing offspring with associated expenditures, such as resource acquisition and predator avoidance. Nest success is an important parameter underlying avian population dynamics. Galliforms are particularly susceptible to low nest success due to exposure of ground nests to multiple predator guilds, lengthy incubation periods, and substantive reliance on crypsis for survival. Hence, it is plausible that nesting individuals prioritize productivity and survival differently, resulting in a gradient of reproductive strategies. Fine-scale movement patterns during incubation are not well documented in ground-nesting birds, and the influence of reproductive movements on survival is largely unknown. Using GPS data collected from female wild turkeys (<i>n</i> = 278) across the southeastern United States, we evaluated the influence of incubation recess behaviors on tradeoffs between nest and female survival. We quantified daily recess behaviors including recess duration, recess frequency, total distance traveled, and incubation range size for each nest attempt as well as covariates for nest concealment, nest attempt, and nest age. Of 374 nests, 91 (24%) hatched and 39 (14%) females were depredated during incubation. Average nest survival during the incubation period was 0.19, whereas average female survival was 0.78. On average, females took 1.6 daily unique recesses (SD = 1.2), spent 2.1 hrs off the nest each day (SD = 1.8), and traveled 357.6 m during recesses (SD = 396.6). Average nest concealment was 92.5 cm (SD = 47). We found that females who took longer recess bouts had higher individual survival, but had increased nest loss. Females who recessed more frequently had lower individual survival. Our findings suggest behavioral decisions made during incubation represent life-history tradeoffs between predation risk and reproductive success on an unpredictable landscape.</p>
Metadata model with controlled vocabularies proposed for the Social Sciences domain
<p>This dataset is related to the MSc work "Descrição de dados de investigação: requisitos de investigadores para modelos de metadados na Psicologia e Ciências da Educação", where, by observing practices, identifying needs, knowing work processes and identifying expectations, and trying to take advantage of researchers' knowledge about the importance of research data in their activity, we have analyzed procedures for accessing, storing and preserving research data. We also have studied models and workflows for managing datasets of completed projects. Considering the recommendations of Data Curation, we proposed the description model that was later expanded with specific examples of controlled vocabularies used in the Psychology scientific domain.</p>
Nest survival, female survival, and average covariates used in modeling, along with metadata
Open the record for dataset details and reuse information.
Metadata Enrichment of Long Text Documents using Large Language Models
<p><span lang="EN-US"><span>In this project, we semantically enriched and enhanced the metadata of long text documents, </span><span>theses</span><span> and dissertations, retrieved from the </span><span>HathiTrust</span><span> Digital Library in English published from 1920 to 2020 through a combination of manual efforts and large language models. This dataset provides a valuable resource for advancing research in areas such as computational social science, digital humanities, and information science. Our paper shows that enriching metadata using LLMs is particularly beneficial for digital repositories by introducing </span><span>additional</span><span> metadata access points that may not have originally been </span><span>foreseen</span><span> to accommodate various content types. This approach is particularly effective for repositories that have significant missing data in their existing metadata fields, enhancing search results, and improving the accessibility of the digital repository.</span></span><span> </span></p>
Malicious Large Language Models Detection using Metadata Information
<p># Introduction<br># This is the replication package for the paper "Malicious Large Language Models Detection using Metadata Information".</p> <p><br>## Task Definition<br>Given the information of LLM, the task is to identify whether it is a malicious LLM that may attack software systems. We treat the task as binary classification (0/1), where 1 stands for malicious LLMs and 0 for malicious-free LLMs.</p> <p><br>## The static of dataset<br>### Data Format<br>Before preprocessing dataset, each line in the uncompressed file represents multiple metadata of one large language model (LLM). One row is illustrated below.<br> - **idx:** the index of example<br> - **repo_id:** the id of LLM (e.g., microsoft/codebert-base)<br> - **tags:** the tags of LLM<br> - **pipeline_tag:** the pipeline_tag of LLM<br> - **downloads:** the number of downloads<br> - **created_time:** the created time of LLM<br> - **modelCard:** the text content of LLM<br> - **num_discussion:** the number of discussions<br> - **discussion:** the discussions of LLM<br> - **para_size:** the size of LLM<br> - **tensor_type:** the type of LLM<br> - **num_commit:** the number of commits<br> - **commit:** the commit of LLM</p> <p>After preprocessing dataset, you can obtain three .csv files, i.e. train.csv, valid.csv, test.csv<br> - **idx:** the index of example.<br> - **repo_id:** the id of LLM (e.g., microsoft/codebert-base).<br> - **tags:** the tags of LLM.<br> - **pipeline_tags:** the pipeline_tag of LLM.<br> - **created_time:** the created time of LLM.<br> - **model_size:** the size of LLM.<br> - **Tensor_type:** the type of LLM.<br> - **is_model_card:** Whether to include model card. If the model has model card, the value is 1, otherwise, the value is 0.<br> - **malicious_model_card:** Whether to include keywords describing the malicious model in model card. If the model card has malicious keywords, the value is 1, otherwise, the value is 0<br> - **repository_link:** Whether to include repository link: GitHub link, Arxiv link, homepage link, bugs link and issues link in model card. If the model card has link, the value is 1, otherwise, the value is 0.<br> - **dataset_info:** Whether to include the adopted dataset information in model card. If the model card has dataset information, the value is 1, otherwise, the value is 0.<br> - **metrics_info:** Whether to include the evaluated metrics information in model card. If the model card has evaluation metrics information, the value is 1, otherwise, the value is 0.<br> ?- **script_info:** Whether to include script information. If the model has script information, the value is 1, otherwise, the value is 0.<br> - **config_content:** The content of the configuration script file. This value is string type.<br> - **stakeholder_name:** The name of authors, contributors, and maintainers. This value is string type.<br> - **number_discussion:** The number of discussion. <br> - **num_pr:** The number of pull request.<br> - **malicious_discussion:** Whether the discussion contains malicious behavior keywords. If the discussion has malicious behavior keywords, the value is 1, otherwise, the value is 0.<br> - **number_commit:** The number of commit.<br> - **malicious_commit:** Whether the title and message of commits contain malicious behavior keywords. If the commit has malicious behavior keywords, the value is 1, otherwise, the value is 0.<br> - **z_download:** The z-score of number of download.<br> - **z_like:** The z-score of number of likes.</p> <p><br>### Data Statistics<br>Data statistics of the dataset are shown in the below table:</p> <p>| #File Names | #Examples |<br>| ------------------- | :------------------------: |<br>| dataset_feature.csv | 578,502 (560,257/18,245) | <br>| train_imbalance.csv | 462,801 (448205/14596) |<br>| valid_imbalance.csv | 57,849 (56025/1824) |<br>| test_imbalance.csv | 57,852 (56027/1825) |<br>| train_balance.csv | 29192 (14596/14596) |<br>| valid_balance.csv | 3648 (1824/1824) |<br>| test_balance.csv | 3650 (1825/1825) |<br>| train_imbalance_50.csv | 289252 (280129/9123) |<br>| train_imbalance_60.csv | 347101 (336154/10947) |<br>| train_imbalance_70.csv | 404952 (392180/12772) |</p> <p>29646 models contain github link<br>all_dataset.csv 596383<br>all_dataset_information.csv 589140 (safe_dataset.csv(570549), unsafe_dataset.csv(18591))<br>safe_dataset_information.csv (559582), unsafe_dataset_information.csv (18212)</p> <p>Description Feature: 'malicious_model_card', 'repository_link', 'dataset_info', 'metrics_info', 'config_conteng'<br>Stakeholder Feature: 'stakeholder_name'<br>Event Feature: 'num_pr', 'number_commit', 'malicious_commit',<br>Context Feature: 'z_download', 'z_like'</p> <p><br>## Pipeline-MPTMHunter<br>We also provide a pipeline that fine-tunes [MPTMHunter](https://doi.org/10.5281/zenodo.12578531) on this task.</p> <p>### Experimental environment configuration<br>```bash<br>huggingface_hub 0.23.1<br>libxgboost 2.0.3<br>lightgbm 4.3.0<br>networkx 3.2.1<br>nltk 3.8.1<br>numpy 1.26.3<br>openssl 3.0.13<br>pandas 2.2.1<br>pillow 10.2.0<br>scikit-learn 1.4.2<br>scipy 1.13.0<br>torch 2.3.0+cu118<br>torchaudio 2.3.0+cu118<br>torchvision 0.18.0+cu118<br>tqdm 4.66.2<br>transformers 4.37.2<br>xgboost 2.0.3<br>```</p> <p>### Dataset Collection Script<br>```bash<br>python ./script/DataExtraction.ipynb<br>python ./script/dataset_spider.py<br>python ./script/config_crawl.py<br>```</p> <p>### Dataset Preprocess Script<br>```bash<br>python feature_generation.py --input_file='../dataset/dataset_information.csv' --output_file='../dataset/dataset_feature.csv'</p> <p>python feature_generation.py --input_file='../dataset/real_world_dataset_information_0701.csv' --output_file='../dataset/real_world_dataset_feature_0701.csv'<br>```</p> <p>### Model Training Script<br>```bash<br>python run_codet5_lstm.py --output_dir='../saved_models/codet5_lstm_imbalance_final' --model_type=codet5 --tokenizer_name='../models/codet5' --model_name_or_path='../models/codet5' --do_train --train_data_file='../dataset/train_imbalance_70.csv' --eval_data_file='../dataset/valid_imbalance_70.csv' --test_data_file='../dataset/test_imbalance.csv' --epoch=3 --block_size=510 --train_batch_size=64 --eval_batch_size=64 --learning_rate=2e-5 --max_grad_norm=1.0 --evaluate_during_training --seed=123456<br>```</p> <p>### Model Inference Script<br>```bash<br>python run_codet5_lstm.py --output_dir='../saved_models/codet5_lstm_imbalance' --model_type=codet5 --tokenizer_name='../models/codet5' --model_name_or_path='../models/codet5' --do_eval --do_test --train_data_file='../dataset/train_imbalance_70.csv' --eval_data_file='../dataset/valid_imbalance_70.csv' --test_data_file='../dataset/test_imbalance.csv' --epoch=3 --block_size=510 --train_batch_size=64 --eval_batch_size=64 --learning_rate=2e-5 --max_grad_norm=1.0 --evaluate_during_training --seed=123456<br>```</p> <p>### Evaluation Script<br>```bash<br>python ../evaluation/evaluation.py -a ../dataset/test_balance.csv -p ../saved_models/codebert_imbalance_all/predictions.txt<br>python ../evaluation/evaluation.py -a ../dataset/test_imbalance.csv -p ../dataset/predictions.txt<br>```</p> <p>## Result<br>The results on the test set are shown as below (We use the OpenTextClassification as the baseline):</p> <p>| Methods | ACC | Precision | Recall | F1-Score |<br>| Random Forest | 97.18% | 83.80% | 13.04% | 22.57% |<br>| LR | 96.80% | 43.46% | 4.55% | 8.23% |<br>| LightGBM | 97.62% | 95.18% | 25.97% | 40.81% |<br>| TextRNN | 96.86% | 73.33% | 0.60% | 1.20% |<br>| TextCNN | 98.89% | 95.61% | 68.00% | 79.47% |<br>| TextRCNN | 98.87% | 95.36% | 67.62% | 79.13% |<br>| TextRNN_Att | 98.94% | 95.29% | 69.86% | 80.62% |<br>| MPTMHunter | ** 99.99% ** | ** 99.95% ** | ** 99.78% ** | ** 99.86% ** |</p>
Human-LLM experiment for the paper: "Towards Generative AI-driven Metadata Modelling: A Human-LLM Collaborative Approach". Generated by Mayukh Bagchi on 1st June, 2024.
Open the record for dataset details and reuse information.
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.