{"id":"W2974657091","doi":"10.4018/ijossp.2019070103","title":"A Topic Modeling Based Approach for Enhancing Corpus Querying","year":2019,"lang":"en","type":"article","venue":"International Journal of Open Source Software and Processes","topic":"Advanced Text Analysis Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Computer science; Query expansion; Ranking (information retrieval); Information retrieval; Query optimization; Process (computing); Selection (genetic algorithm); Web query classification; Web search query; Data mining; Query language; Sargable; Quality (philosophy); Search engine; Machine learning","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":[],"consensus_categories":[],"category_scores_codex":[0.003698682,0.001251312,0.001312971,0.00391326,0.001243051,0.003169592,0.001825298,0.001394313,0.00249638],"category_scores_gemma":[0.01123547,0.0006022961,0.001862507,0.005307352,0.000873923,0.005439719,0.002643154,0.001789814,0.001757736],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009262008,"about_ca_system_score_gemma":0.001551559,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005218955,"about_ca_topic_score_gemma":0.005775075,"domain_scores_codex":[0.9966228,0.001278577,0.0002989319,0.0007115548,0.0009310023,0.0001571834],"domain_scores_gemma":[0.9952657,0.002560673,0.0002174838,0.0007577986,0.001076905,0.0001213538],"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":[0.0007443586,0.0004239123,0.00390739,0.001020344,0.0003495704,0.000613953,0.003345473,0.03672868,0.09143879,0.0527238,0.02563731,0.7830665],"study_design_scores_gemma":[0.0001063625,0.0003328686,0.00251762,0.00008602324,0.0002793007,0.001087871,0.0009231839,0.8379526,0.04035776,0.04875599,0.06739199,0.0002085062],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007347829,0.001001055,0.9849756,0.000336343,0.00009901931,0.0002306672,0.0004311223,0.0038904,0.001687975],"genre_scores_gemma":[0.1353414,0.001304264,0.8549154,0.0003309676,0.0003414913,0.00061425,0.002283221,0.0009171915,0.003951774],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005218955,"threshold_uncertainty_score":0.01956069,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02535517324778782,"score_gpt":0.3171102658473063,"score_spread":0.2917550925995185,"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."}}