{"id":"W4200630255","doi":"10.1093/bib/bbac279","title":"Implications of topological imbalance for representation learning on biomedical knowledge graphs","year":2022,"lang":"en","type":"preprint","venue":"Briefings in Bioinformatics","topic":"Bioinformatics and Genomic Networks","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute; Canadian Institute for Advanced Research; HEC Montréal","funders":"","keywords":"Inference; Computer science; Embedding; Theoretical computer science; Domain knowledge; Graph embedding; Representation (politics); Graph; Knowledge graph; Topological graph theory; Machine learning; Biological network; Artificial intelligence; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005205465,0.0002469067,0.0003740672,0.0001803579,0.0001297887,0.00003324604,0.0004745874,0.0004621507,0.00002712307],"category_scores_gemma":[0.0002695971,0.0002480915,0.0002293702,0.0001948907,0.0001771793,0.000005408478,0.0007690354,0.0005586322,0.000002586569],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005443155,"about_ca_system_score_gemma":0.0001730045,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002198774,"about_ca_topic_score_gemma":0.000009660798,"domain_scores_codex":[0.9982396,0.00004803598,0.0009462967,0.0002881536,0.0001614843,0.000316434],"domain_scores_gemma":[0.9985766,0.0001066283,0.0006240803,0.0005167848,0.0001024415,0.00007345165],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.002317847,0.004305837,0.04031165,0.01038854,0.001526443,0.000006792358,0.01826331,0.1271441,0.01436123,0.2833097,0.2436249,0.2544397],"study_design_scores_gemma":[0.008685997,0.006115363,0.04812816,0.001120716,0.000296071,0.0000856765,0.004236122,0.2360557,0.007483152,0.1184369,0.5654063,0.003949801],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7342843,0.003703139,0.205071,0.005098571,0.002895225,0.006798183,0.001908296,0.0001937148,0.04004755],"genre_scores_gemma":[0.9546279,0.001530327,0.03566588,0.001116943,0.0002301713,0.0006100114,0.005717464,0.0000525631,0.0004487652],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3217814,"threshold_uncertainty_score":0.9999971,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02507420569848868,"score_gpt":0.3141767155849239,"score_spread":0.2891025098864353,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}