{"id":"W2905726793","doi":"10.22148/16.030","title":"Critical Search: A Procedure for Guided Reading in Large-Scale Textual Corpora","year":2018,"lang":"en","type":"article","venue":"Journal of Cultural Analytics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Reading (process); Computer science; Scale (ratio); Information retrieval; Natural language processing; Artificial intelligence; Linguistics; Geography; Philosophy; Cartography","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":[0.02680897,0.003155011,0.002089575,0.01796972,0.00532086,0.006743511,0.003541521,0.002251031,0.06849349],"category_scores_gemma":[0.1220755,0.00238242,0.002421336,0.0134338,0.004461799,0.00823401,0.01011567,0.00359151,0.02723692],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002258291,"about_ca_system_score_gemma":0.009621345,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004828549,"about_ca_topic_score_gemma":0.01235764,"domain_scores_codex":[0.980667,0.009337437,0.003499361,0.003136174,0.002910244,0.000449863],"domain_scores_gemma":[0.9090465,0.06456757,0.003455837,0.01029953,0.01144255,0.001188052],"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":[0.0007503851,0.0002468888,0.001907921,0.005095237,0.0003074666,0.001212249,0.02700516,0.001767977,0.01591217,0.08144737,0.2851336,0.5792136],"study_design_scores_gemma":[0.0005189066,0.0002040883,0.003441037,0.001802304,0.0002422044,0.001276067,0.01696797,0.03772385,0.02307173,0.2710851,0.6430191,0.0006477128],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004506723,0.0003631263,0.9176248,0.001276471,0.0004470203,0.005114101,0.01255363,0.0433385,0.01477557],"genre_scores_gemma":[0.01513982,0.0001850091,0.9639943,0.0002735906,0.0001293761,0.004572228,0.006637223,0.005000161,0.004068202],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.06849349,"threshold_uncertainty_score":0.2291335,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0522741677462495,"score_gpt":0.3675912135698836,"score_spread":0.3153170458236341,"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."}}