{"id":"W2359946253","doi":"","title":"Lightweight Document clustering Algorithm for Forum FAQ Document","year":2008,"lang":"en","type":"article","venue":"Microcomputer applications","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Document clustering; Cluster analysis; Information retrieval; Data mining; Algorithm; Artificial intelligence","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"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":[0.00118074,0.0007038201,0.001061308,0.004400474,0.001911105,0.001769647,0.00188387,0.001381649,0.007904965],"category_scores_gemma":[0.004886096,0.0002732659,0.0009368874,0.004559054,0.0004622822,0.001837051,0.00132445,0.001042085,0.008078814],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001261248,"about_ca_system_score_gemma":0.002563927,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007018961,"about_ca_topic_score_gemma":0.005996739,"domain_scores_codex":[0.9983967,0.0002030846,0.0001332045,0.0003910864,0.0007208741,0.0001550834],"domain_scores_gemma":[0.9981198,0.000314432,0.0001455118,0.0004008513,0.0009287793,0.00009063012],"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":[0.0002667136,0.0001098407,0.001270729,0.0001920505,0.00005980782,0.000138232,0.0001695062,0.03627138,0.01465735,0.01545119,0.03665641,0.8947567],"study_design_scores_gemma":[0.0001077861,0.0001421053,0.0018738,0.0000498008,0.0000445123,0.0007128963,0.0001888424,0.8556177,0.01942185,0.0340606,0.08767468,0.000105472],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007242327,0.0005116698,0.9823735,0.0002395307,0.0002117608,0.0003302587,0.001139989,0.005132683,0.002818226],"genre_scores_gemma":[0.06388463,0.00033514,0.9183146,0.0001147369,0.0001744737,0.0005460818,0.004150991,0.0003375058,0.01214196],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007904965,"threshold_uncertainty_score":0.02644479,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01194599614869509,"score_gpt":0.2437200610914195,"score_spread":0.2317740649427244,"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."}}