{"id":"W2096794280","doi":"10.1109/ccece.2011.6030580","title":"Classifying the Arabic web — A pilot study","year":2011,"lang":"en","type":"article","venue":"","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Computer science; World Wide Web; Arabic; Web crawler; Web page; Globe; Categorization; The Internet; Information retrieval; Artificial intelligence; 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":[],"consensus_categories":[],"category_scores_codex":[0.0002347347,0.00007694658,0.00006457556,0.00006528183,0.0001562737,0.0001188681,0.001438522,0.00001453789,0.0001252297],"category_scores_gemma":[0.0000285646,0.00004223169,0.00001939085,0.0003386982,0.00005687408,0.0003363285,0.0003277282,0.0000985333,0.0002916154],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001345532,"about_ca_system_score_gemma":0.00002548515,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003490128,"about_ca_topic_score_gemma":0.0000507386,"domain_scores_codex":[0.9992852,0.00003833273,0.0001412956,0.0002261117,0.0001553668,0.0001536797],"domain_scores_gemma":[0.9990339,0.00004328213,0.00005189211,0.0008201283,0.00002876656,0.00002207522],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.000006319775,0.0007918309,0.02161515,0.000001830215,0.00002272043,0.000005348795,0.001980123,2.963867e-7,0.0008364536,0.9212191,0.005629296,0.04789156],"study_design_scores_gemma":[0.002534659,0.008386672,0.7064978,0.00001818383,0.00004307765,0.00003205629,0.018951,0.02113893,0.01282281,0.1804477,0.04786476,0.001262315],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3063544,0.0001413323,0.4134763,0.008492771,0.0007581174,0.001175112,3.926903e-7,0.005367936,0.2642336],"genre_scores_gemma":[0.9915032,0.000005245174,0.006496481,0.0002166192,0.000009167078,0.00005513333,6.455848e-8,0.000003213619,0.001710845],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7407714,"threshold_uncertainty_score":0.3748223,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1525639738533988,"score_gpt":0.272155169388718,"score_spread":0.1195911955353193,"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."}}