{"id":"W4229010387","doi":"10.16995/dscn.8106","title":"Concept Detection in Philosophical Corpora","year":2022,"lang":"en","type":"article","venue":"Digital Studies / Le champ numérique","topic":"Advanced Text Analysis Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Cosine similarity; Natural language processing; Document retrieval; Information retrieval; Similarity (geometry); Artificial intelligence; Word (group theory); Embedding; Linguistics; Pattern recognition (psychology); Philosophy; Image (mathematics)","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002250008,0.0001938904,0.0003344553,0.0001874887,0.0002926128,0.0000879666,0.000647708,0.00004095789,0.000004573328],"category_scores_gemma":[0.0001252684,0.0002021691,0.0001160439,0.00093601,0.0001368806,0.0008170517,0.001297363,0.0003534849,0.000008025851],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003229836,"about_ca_system_score_gemma":0.00004253142,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002067316,"about_ca_topic_score_gemma":0.00005152815,"domain_scores_codex":[0.998402,0.00009125641,0.0003350852,0.0005296886,0.000341103,0.0003008552],"domain_scores_gemma":[0.9991572,0.0001262429,0.0001479057,0.0004238512,0.00008881617,0.00005595072],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0002013219,0.00284549,0.007454121,0.0001070647,0.0008246853,0.001714754,0.398585,0.00890175,0.005331435,0.09164695,0.003035557,0.4793519],"study_design_scores_gemma":[0.001038017,0.0006702849,0.000499884,0.00002846062,2.505038e-7,0.0001257015,0.07694628,0.007398155,0.01363851,0.8802648,0.01839021,0.0009994187],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3199833,0.001722961,0.6625121,0.007302358,0.0004200128,0.0006036939,0.00002435868,0.001214544,0.006216642],"genre_scores_gemma":[0.9988986,0.00004489709,0.0002851221,0.0002991314,0.00006134211,0.0002507805,0.000007814532,0.0000156722,0.0001366789],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7886178,"threshold_uncertainty_score":0.8244215,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02500027851965196,"score_gpt":0.2731868783958897,"score_spread":0.2481865998762377,"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."}}