{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001230707,0.001450632,0.001562588,0.005012567,0.001650367,0.009452282,0.002353815,0.00162704,0.8537714],"category_scores_gemma":[0.01457317,0.0007611397,0.000988118,0.006280774,0.0007878461,0.005373264,0.003839119,0.00276769,0.8540971],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002337138,"about_ca_system_score_gemma":0.003179286,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006282926,"about_ca_topic_score_gemma":0.006240529,"domain_scores_codex":[0.9982332,0.0001763648,0.0001202332,0.0003509547,0.0009759385,0.0001433841],"domain_scores_gemma":[0.9931726,0.001421758,0.0002557536,0.001337754,0.003210416,0.0006017218],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00002044227,0.000009368197,0.00003472098,0.0001376219,0.000001917972,0.000009954303,0.00001613859,0.00002689333,0.00009462555,0.001665642,0.9624136,0.03556896],"study_design_scores_gemma":[0.000007327412,0.000003810518,0.000102176,0.00007328122,0.000001746382,0.000008897094,0.00002047943,0.00002830318,0.00009823529,0.001013128,0.9986388,0.00000385777],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"other","genre_gemma":"methods","genre_scores_codex":[0.0003107955,0.001117773,0.003183657,0.005392922,0.009771827,0.0004214245,0.09057466,0.00554617,0.8836808],"genre_scores_gemma":[0.001114563,0.0006828764,0.001223589,0.001941461,0.0009472386,0.0002484856,0.02689243,0.002709998,0.9642394],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8537714,"threshold_uncertainty_score":0.2085774,"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."}}