{"id":"W2029246745","doi":"10.1002/meet.2008.1450450271","title":"Approximate phrase searching: Movie scripts and song lyrics","year":2008,"lang":"en","type":"article","venue":"Proceedings of the American Society for Information Science and Technology","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Phrase; Phrase search; Computer science; Scripting language; Complement (music); Information retrieval; Task (project management); Lyrics; Search engine; Full text search; Natural language processing; Artificial intelligence; Web search query; Search analytics","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":"codex-gemma-dda1882f352a","candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.0008494651,0.00008198425,0.0001592608,0.0002437375,0.0007198536,0.0001396545,0.0009902415,0.00002988582,1.009137e-7],"category_scores_gemma":[0.0003805314,0.00005870287,0.00005750929,0.003151808,0.002785243,0.002722946,0.0006854543,0.0001067118,5.837736e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002626698,"about_ca_system_score_gemma":0.0001175135,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000148993,"about_ca_topic_score_gemma":1.173248e-7,"domain_scores_codex":[0.9989882,0.000001173356,0.0001985824,0.0001871037,0.0003563777,0.0002686027],"domain_scores_gemma":[0.9988686,0.00002118392,0.0003460704,0.0001577733,0.0005517158,0.00005469548],"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":[0.0000181965,0.00007108302,0.04077036,0.0002733258,0.0000888795,1.793521e-7,0.0193513,0.000008142667,0.05210903,0.2839806,0.01019286,0.5931361],"study_design_scores_gemma":[0.001770683,0.001074356,0.01718225,0.0001387512,0.0001042397,0.0004806443,0.04939064,0.7630181,0.09193346,0.0229777,0.0507837,0.001145415],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9836293,0.00003378406,0.01163726,0.004059222,0.00002383836,0.0001490326,0.000007246398,0.0001126843,0.0003476328],"genre_scores_gemma":[0.9398271,0.0001755528,0.05942841,0.0005311248,0.000005881275,0.00001614236,5.052953e-7,0.000002087756,0.0000131743],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.76301,"threshold_uncertainty_score":0.9999286,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01721096691679572,"score_gpt":0.2494692875875509,"score_spread":0.2322583206707551,"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."}}