{"id":"W830770281","doi":"10.4000/trajectoires.2157","title":"Construction de corpus généraux et spécialisés à partir du Web (Ad hoc and general-purpose corpus construction from web sources)","year":2016,"lang":"fr","type":"article","venue":"Tr@jectoires","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Barrie Urology Group","funders":"","keywords":"Humanities; Corpus linguistics; Art; Philosophy; Linguistics","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005629113,0.0005893694,0.0006140686,0.000271642,0.0004870787,0.0007827072,0.0007469476,0.0005413811,0.0002023475],"category_scores_gemma":[0.0004084074,0.0005029244,0.0001617811,0.0005478711,0.001823272,0.001584684,0.0004734288,0.0004892514,0.00006360546],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003494792,"about_ca_system_score_gemma":0.000802617,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001365383,"about_ca_topic_score_gemma":0.001037236,"domain_scores_codex":[0.9961995,0.0006146607,0.000678846,0.001122115,0.0005293408,0.000855487],"domain_scores_gemma":[0.9975057,0.0004896554,0.0005966399,0.0007430156,0.0003063122,0.0003587048],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001609034,0.00008780343,0.01775728,0.00008038151,0.000086338,0.0001750567,0.001360226,0.000003580675,0.1279124,0.06695943,0.003789943,0.7816267],"study_design_scores_gemma":[0.006330993,0.0009886442,0.008224203,0.002652629,0.0004913353,0.005551584,0.0003350477,0.0184213,0.3707694,0.3514989,0.2316582,0.003077767],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8641513,0.05835242,0.05692758,0.01395217,0.004019256,0.0004814584,0.0002923769,0.001466604,0.0003568148],"genre_scores_gemma":[0.6224813,0.01145271,0.3620864,0.0009757263,0.00133779,0.00005777475,0.00002389531,0.00008104013,0.001503391],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7785489,"threshold_uncertainty_score":0.9997422,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01616006285433942,"score_gpt":0.255013822750617,"score_spread":0.2388537598962775,"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."}}