{"id":"W2561412020","doi":"10.1371/journal.pone.0181142","title":"\"What is relevant in a text document?\": An interpretable machine learning approach","year":2017,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":268,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institute for Information and Communications Technology Promotion; Banting and Best Diabetes Centre, University of Toronto; National Research Foundation of Korea; Deutsche Forschungsgemeinschaft; National Research Foundation","keywords":"Computer science; Artificial intelligence; Relevance (law); Natural language processing; Categorization; Support vector machine; Convolutional neural network; Classifier (UML); Word (group theory); Machine learning; Document classification; Tracing; Information retrieval","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001784711,0.001197881,0.0005494407,0.002244995,0.0004556582,0.002738279,0.001942799,0.001538145,0.002971648],"category_scores_gemma":[0.009383699,0.0004415542,0.0007649055,0.001426302,0.00159104,0.003605742,0.001135869,0.001883863,0.0008334924],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00101504,"about_ca_system_score_gemma":0.0006270703,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0019639,"about_ca_topic_score_gemma":0.001818345,"domain_scores_codex":[0.9989937,0.0004858221,0.00005682249,0.000250178,0.0001770241,0.00003656347],"domain_scores_gemma":[0.9968724,0.002166874,0.0003660162,0.0002717061,0.0002630381,0.00006008455],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004421533,0.0003191946,0.01437016,0.001503667,0.0003627239,0.001819949,0.004806565,0.08225408,0.02561693,0.4511248,0.02017825,0.3972015],"study_design_scores_gemma":[0.00003036907,0.00007980037,0.004025928,0.0002400612,0.000108977,0.0006777484,0.0006698153,0.4920511,0.003655209,0.4760374,0.02237761,0.0000460556],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03734449,0.002776199,0.9390346,0.009319632,0.0002773799,0.000144433,0.001927835,0.001377854,0.007797601],"genre_scores_gemma":[0.6194977,0.002210054,0.3692596,0.00120322,0.0006493824,0.0002529364,0.002173818,0.0002472472,0.004506122],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002971648,"threshold_uncertainty_score":0.009941161,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06341091681322418,"score_gpt":0.2827673665988876,"score_spread":0.2193564497856634,"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."}}