{"id":"W4388975825","doi":"10.1007/978-3-031-45043-3_4","title":"Text to Data","year":2023,"lang":"en","type":"book-chapter","venue":"Synthesis lectures on data management","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Natural language processing; Parsing; Artificial intelligence; Principle of compositionality; Semantic computing; Graph; Embedding; Information retrieval; Semantic Web; Theoretical computer science","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":["metaepi_narrow","open_science","insufficient_payload"],"consensus_categories":["open_science"],"category_scores_codex":[0.0008580401,0.000511562,0.000444784,0.0005706419,0.0001439727,0.0005008041,0.01947005,0.0001719029,0.0002522136],"category_scores_gemma":[0.0002149638,0.0004726876,0.00006346708,0.0001281462,0.00002689194,0.0004098386,0.02317393,0.000331479,0.004567766],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009562789,"about_ca_system_score_gemma":0.00004176465,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003167426,"about_ca_topic_score_gemma":0.0001102499,"domain_scores_codex":[0.9952638,0.00003974884,0.0004395523,0.002778443,0.001004327,0.000474132],"domain_scores_gemma":[0.9800651,0.0003471374,0.0001545209,0.01924837,0.00002640936,0.0001584741],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000004990467,0.00001112977,9.547806e-8,0.00007534486,0.0002086336,0.00012907,0.00001880467,0.0002217827,7.672029e-7,0.2382835,0.2797703,0.4812756],"study_design_scores_gemma":[0.0000558244,0.00002205689,0.0000123935,0.0004152665,0.0001053411,0.000002328262,0.000003665211,0.01973309,0.00001465796,0.01449313,0.9645804,0.0005619163],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"other","genre_gemma":"other","genre_scores_codex":[2.830601e-7,0.0001542365,0.376587,0.005661224,0.001126062,0.0008220975,0.001023855,0.0006635952,0.6139616],"genre_scores_gemma":[0.0006739869,0.0008471941,0.1333309,0.00619732,0.000922218,0.0000983762,0.001706156,0.000267601,0.8559563],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.68481,"threshold_uncertainty_score":0.9997725,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1342594939711177,"score_gpt":0.2972451479885176,"score_spread":0.1629856540173999,"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."}}