AI4PROFHEALTH - Profession-health status co-occurrence graph statistics

<p>This dataset contains the Pointwise Mutual Information (PMI) values for co-occurrence pairs between different mention categories extracted from two distinct clinical datasets: <strong>MESINESP2</strong> and the <strong>Clinical Case Reports Collection</strong>. PMI is a statistical measure used to assess the strength of association between pairs of entities by comparing their observed co-occurrence to the expected frequency under the assumption of independence.</p> <p>The datasets include PMI values for each co-occurrence pair, derived from the association of professions and clinical concepts, with the aim of identifying potential occupational health risks.&nbsp;By sharing these datasets, we aim to support further research into the relationships between professions and clinical entities, enabling the development of more accurate and targeted occupational health risk models.</p> <p>There is a separate file for each corpus, and each dataset is provided in <strong>CSV format</strong> for easy access and analysis. These files include the PMI values for co-occurrence pairs extracted from the respective corpora, making them suitable for further data analysis.</p> <p><strong>Data Structure:</strong></p> <ul> <li><strong>MESINESP2: <code>mesinesp2_co-occurrence_pmi.zip</code></strong></li> <li><strong>Clinical case reports:&nbsp;<code>clinical_cases_co-occurrence_pmi.zip</code></strong></li> </ul> <p>The repository contains a .zip file for each of the corpus, each containing a .csv file with the co-occurrences between the detected professions and clinical entities. The file has the following columns order:</p> <ul> <li><code><strong>span_mention_1</strong></code>: Mention string (original): profession</li> <li><code><strong>normalized_entity_1</strong></code>: Controlled vocabulary entry for this term</li> <li><code><strong>mention1_category</strong></code>: Semantic class (i.e., NER label)</li> <li><code><strong>mention1_freq</strong></code>: Absolute frequency of this mention entity 1</li> <li><code><strong>span_mention_2</strong></code>:<strong> </strong>Mention string (original): entity 2 (disease, symptom, species, etc.)</li> <li><code><strong>normalized_entity_2</strong></code>: Controlled vocabulary entry for this term</li> <li><code><strong>mention2_category</strong></code>:<strong> </strong>Semantic class (i.e., NER label)</li> <li><code><strong>mention1_freq</strong></code>:<strong> </strong>Absolute frequency of this mention entity 2</li> <li><code><strong>co-occurrence</strong></code>: Number of co-occurrences</li> <li><code><strong>PMID</strong></code>: PMID value</li> </ul> <p><strong>Notes</strong></p> <p>This resource been funded by the Spanish National Proyectos I+D+i 2020 AI4ProfHealth project PID2020-119266RA-I00 (PID2020-119266RA-I0/AEI/10.13039/501100011033).</p> <p><strong>Contact</strong></p> <p>If you have any questions or suggestions, please contact us at:</p> <p>- Miguel Rodr&iacute;guez Ortega (&lt;miguel [dot] rod [at] bsc [dot] com&gt;)<br>- Martin Krallinger (&lt;krallinger [dot] martin [at] gmail [dot] com&gt;)</p> <p><strong>Additional resources and corpora</strong></p> <p>If you are interested, you might want to check out these corpora and resources:</p> <ul> <li><a href="../records/7116201">MEDDOPROF</a> (Corpus of mentions of professions, occupations and working status and normalization, different document collection with some overlapping documents)</li> <li><a href="https://zenodo.org/records/5602914">MESINESP-2</a>&nbsp;(Corpus of manually indexed records with DeCS /MeSH terms comprising scientific literature abstracts, clinical trials, and patent abstracts, different document collection)</li> </ul> <p>&nbsp;</p>