{"id":"W4410512285","doi":"10.1007/978-3-031-51447-0_155-1","title":"Wordlists and Data-Driven Learning","year":2025,"lang":"en","type":"book-chapter","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université du Québec à Montréal","funders":"","keywords":"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":[],"consensus_categories":[],"category_scores_codex":[0.001083917,0.0008028552,0.0007103447,0.001863479,0.0004401137,0.002289995,0.001816714,0.0008168587,0.02122503],"category_scores_gemma":[0.005061807,0.0005311559,0.0007590327,0.003562264,0.0009327254,0.00555226,0.001105345,0.001680089,0.01137565],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001001075,"about_ca_system_score_gemma":0.000908131,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001762668,"about_ca_topic_score_gemma":0.002515584,"domain_scores_codex":[0.9993284,0.0001384367,0.00005662678,0.0001700633,0.0002751277,0.00003149487],"domain_scores_gemma":[0.9979183,0.001420028,0.00005311662,0.0002430005,0.0003263424,0.00003912912],"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":[0.00004101056,0.00004174363,0.0001755407,0.0003063365,0.00001839048,0.00003483021,0.00006585396,0.008849214,0.001242565,0.1053728,0.04102821,0.8428235],"study_design_scores_gemma":[0.00001523681,0.00002725199,0.0002365489,0.0001823854,0.00001862537,0.000149704,0.00007219879,0.1198115,0.006448841,0.6966094,0.1763975,0.00003072814],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002096382,0.00539048,0.9646497,0.0009550877,0.0005400066,0.00006490752,0.001328224,0.005186358,0.01978895],"genre_scores_gemma":[0.05197108,0.007734542,0.852331,0.0008080977,0.0004714582,0.000315282,0.008423957,0.002389572,0.07555506],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02122503,"threshold_uncertainty_score":0.07100481,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02302597609306245,"score_gpt":0.2905915697451161,"score_spread":0.2675655936520536,"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."}}