{"id":"W4206897013","doi":"10.1109/dasa53625.2021.9682402","title":"A Moroccan News Articles Dataset (MNAD) For Arabic Text Categorization","year":2021,"lang":"en","type":"article","venue":"2021 International Conference on Decision Aid Sciences and Application (DASA)","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Chicoutimi","funders":"","keywords":"Computer science; Artificial intelligence; Natural language processing; Support vector machine; Naive Bayes classifier; Confusion matrix; Categorization; Context (archaeology); Arabic; Random forest; Gradient boosting; Confusion; Machine learning; Linguistics; Geography","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":[0.0003532027,0.0001509842,0.0001463048,0.0002048271,0.0003432708,0.000928262,0.001125591,0.00007257625,0.0002241778],"category_scores_gemma":[0.0002895751,0.0001306375,0.00004384591,0.0006612779,0.0001765849,0.0007609808,0.0002828519,0.00009065665,0.0001068955],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004416785,"about_ca_system_score_gemma":0.0001695598,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002165941,"about_ca_topic_score_gemma":0.0000517703,"domain_scores_codex":[0.9980342,0.00003154792,0.0003564075,0.0008046048,0.0005665286,0.0002067164],"domain_scores_gemma":[0.9985899,0.0002312963,0.0002022795,0.0005203139,0.0003624119,0.00009379425],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000100522,0.00005784464,0.0003352844,0.000002068816,0.00000485107,7.176464e-7,0.00004013785,0.00008056823,0.003792443,0.663856,0.005283009,0.326537],"study_design_scores_gemma":[0.0008829052,0.0001626949,0.00210577,0.00004128828,0.000007656895,0.00001724652,0.0009003715,0.484062,0.01466127,0.3042429,0.1925482,0.0003676556],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01393387,0.00008000164,0.9537733,0.02918845,0.0002977611,0.0002978705,0.0001254003,0.00009759494,0.002205808],"genre_scores_gemma":[0.9463208,0.0004675199,0.05107859,0.0009674228,0.00006872504,0.0002075332,0.0002880091,0.000006195503,0.0005951638],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.932387,"threshold_uncertainty_score":0.8951252,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06935459005021938,"score_gpt":0.3509310044313946,"score_spread":0.2815764143811752,"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."}}