{"id":"W2120311929","doi":"10.16995/dscn.247","title":"NUMEXCO: A Text Mining Approach to Thematic Analysis of a Philosophical Corpus","year":2005,"lang":"en","type":"article","venue":"Digital Studies / Le champ numérique","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal","funders":"","keywords":"Computer science; Thematic analysis; Software; Thematic map; Modular design; Data science; Corpus linguistics; Natural language processing; Artificial intelligence; Sociology; Qualitative research; Social science; Programming language; Cartography","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003221969,0.0002446988,0.0006770639,0.0003149183,0.0001402422,0.0001538935,0.0006815967,0.0000578241,0.000002391922],"category_scores_gemma":[0.0003371877,0.0001999617,0.0002587035,0.001665399,0.00009139007,0.0003552946,0.0005450233,0.0001658163,0.00002413907],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005085051,"about_ca_system_score_gemma":0.00003598098,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001591587,"about_ca_topic_score_gemma":0.000009769867,"domain_scores_codex":[0.9982625,0.00006280937,0.0004351986,0.0005270141,0.0003703851,0.0003421227],"domain_scores_gemma":[0.9987565,0.0002627001,0.0001693549,0.0005254222,0.0001488703,0.0001371125],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003925659,0.001900372,0.005236341,0.000257538,0.005392427,0.00003028562,0.6895582,0.04363818,0.00009871698,0.0161157,0.001123696,0.2366093],"study_design_scores_gemma":[0.001273061,0.0005659676,0.002647309,0.0002412142,0.00001298323,0.00007379141,0.1058713,0.8687432,0.0003896253,0.01174091,0.006969304,0.001471396],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4969076,0.0007138463,0.4552765,0.007241071,0.0001035417,0.0003009303,0.00001997089,0.000271725,0.03916484],"genre_scores_gemma":[0.9926085,0.00001464807,0.006353605,0.0002910704,0.0001408571,0.00005077425,0.00000934584,0.00001489586,0.0005162521],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.825105,"threshold_uncertainty_score":0.81542,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02825055888610108,"score_gpt":0.2763301992438175,"score_spread":0.2480796403577164,"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."}}