{"id":"W4415121249","doi":"10.1016/j.neucom.2025.131806","title":"FEMTL-DR: A feature-enhanced multi-task learning model for flexible drug recommendation","year":2025,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"Natural Science Foundation of Hubei Province","keywords":"Pairwise comparison; Graph; Node (physics); Feature learning; Adjacency matrix; Artificial neural network; Task (project management); Deep learning; Word embedding","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":[0.000849058,0.0002504152,0.000270037,0.0002860845,0.0005652559,0.0003539506,0.0007603802,0.00006874795,0.000001373697],"category_scores_gemma":[0.000424799,0.0002768348,0.0001586995,0.0007813211,0.00002287544,0.0005279661,0.000578005,0.0004217898,0.000009508772],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000105624,"about_ca_system_score_gemma":0.0002001377,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007282857,"about_ca_topic_score_gemma":0.000003876738,"domain_scores_codex":[0.9978424,0.000275741,0.0003828521,0.0008366692,0.0002065129,0.0004558812],"domain_scores_gemma":[0.9979544,0.001211616,0.0002197594,0.0003243335,0.0002156392,0.00007422632],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001490682,0.00005747345,0.00004580256,0.0000499908,0.00001473163,6.463596e-7,0.0008306278,0.80865,0.002318584,0.0087506,0.001683121,0.1775835],"study_design_scores_gemma":[0.0008350865,0.0000248789,0.0003952429,0.00006423648,0.00000947447,0.000002602841,0.00003291584,0.9846932,0.004752994,0.004706311,0.004252226,0.0002308558],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01274695,0.00003585532,0.9808573,0.003095549,0.0007656082,0.00055672,0.000002888176,0.0007086013,0.001230564],"genre_scores_gemma":[0.416206,0.000003291947,0.5788641,0.001314871,0.00008951945,0.00005759386,0.00002141127,0.00002262641,0.003420576],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4034591,"threshold_uncertainty_score":0.9999684,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03652560952822074,"score_gpt":0.3513498863594664,"score_spread":0.3148242768312456,"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."}}