{"id":"W3082570699","doi":"10.48550/arxiv.2106.14731","title":"The DELICES project: Indexing scientific literature through semantic expansion","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Text Analysis Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Search engine indexing; Computer science; Relevance (law); Information retrieval; Scientific literature; Digital library; Data science; World Wide Web; Linguistics; Political science","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.005447895,0.001496962,0.001338708,0.01725297,0.00134682,0.003475858,0.001304514,0.001129389,0.007535681],"category_scores_gemma":[0.01630667,0.0004974235,0.001533275,0.01241552,0.002011685,0.009864902,0.005902966,0.002017339,0.006192835],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001316335,"about_ca_system_score_gemma":0.003531618,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002799188,"about_ca_topic_score_gemma":0.003211484,"domain_scores_codex":[0.9939151,0.002369471,0.0005974687,0.0009925633,0.001876792,0.0002485125],"domain_scores_gemma":[0.9916788,0.00363503,0.0007501487,0.001742136,0.001751452,0.0004422785],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007661233,0.0004137192,0.001667607,0.003184549,0.0002522285,0.0003468357,0.001588458,0.003681868,0.02402082,0.09960707,0.1320968,0.732374],"study_design_scores_gemma":[0.0005493388,0.0004611519,0.003653755,0.0009610966,0.0003439668,0.0007037739,0.001441994,0.0617405,0.03174008,0.2156914,0.6825021,0.0002108864],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04901604,0.02249993,0.7677624,0.008588606,0.003264685,0.002629731,0.05435422,0.03466374,0.05722062],"genre_scores_gemma":[0.08412324,0.009002444,0.8019573,0.00125341,0.001390433,0.001336167,0.0791132,0.002713313,0.01911054],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9965242,"threshold_uncertainty_score":0.02881157,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06650420550289016,"score_gpt":0.2294048143742566,"score_spread":0.1629006088713665,"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."}}