{"id":"W2182032621","doi":"","title":"Entropy-based Sentence Selection with Roget's Thesaurus","year":2010,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Automatic summarization; Sentence; Computer science; Thesaurus; Information retrieval; Natural language processing; Selection (genetic algorithm); Baseline (sea); Artificial intelligence","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001004806,0.00007239167,0.00005739649,0.00004183898,0.00006655552,0.00009877449,0.0003568843,0.00003208245,0.00005596689],"category_scores_gemma":[0.00001154809,0.00005078202,0.00001926179,0.0001714542,0.00001818327,0.0002292273,0.00003743935,0.0001440812,0.00003975881],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001105041,"about_ca_system_score_gemma":0.00007679657,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009920466,"about_ca_topic_score_gemma":0.0001787216,"domain_scores_codex":[0.9993076,0.00001390412,0.00007822237,0.0002495942,0.0001808113,0.0001698291],"domain_scores_gemma":[0.9994938,0.00002636662,0.00003000295,0.0003296084,0.00006301526,0.00005717203],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002275946,0.0001669396,0.04885899,0.00001673779,0.00002277278,0.0000255315,0.0003735327,0.007452743,0.1193123,0.742479,0.0004455917,0.08082318],"study_design_scores_gemma":[0.0002167389,0.00004916998,0.001161746,0.000004126293,0.000002158058,0.00002506667,0.000006253163,0.9605781,0.03469659,0.001130802,0.002016685,0.0001125859],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1688596,0.000001834785,0.8277669,0.001034807,0.0001620918,0.00006209395,7.81423e-8,0.0002246189,0.001887991],"genre_scores_gemma":[0.7156869,1.604535e-7,0.2836803,0.0003452838,0.00004702029,0.000004928238,1.853494e-7,0.000003107917,0.0002321473],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9531254,"threshold_uncertainty_score":0.207083,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008418135404538249,"score_gpt":0.207359724092659,"score_spread":0.1989415886881207,"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."}}