{"id":"W2885704429","doi":"10.5339/qfarc.2018.ictpd879","title":"Annotation Guidelines for Text Analytics in Social Media","year":2018,"lang":"en","type":"article","venue":"","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Annotation; Modern Standard Arabic; Computer science; Variety (cybernetics); Social media; Arabic; Profiling (computer programming); Analytics; Natural language processing; Linguistics; Artificial intelligence; World Wide Web; Data science","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0568324,0.002524602,0.001863569,0.01744421,0.004296125,0.01070272,0.005531591,0.005396231,0.03185401],"category_scores_gemma":[0.1832471,0.003007751,0.00234238,0.01275911,0.004127095,0.01436505,0.01071939,0.005989955,0.05506568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002787744,"about_ca_system_score_gemma":0.008117534,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008502535,"about_ca_topic_score_gemma":0.01206329,"domain_scores_codex":[0.9281799,0.03211921,0.01735495,0.005545132,0.01520505,0.001595683],"domain_scores_gemma":[0.7202429,0.1173908,0.01243716,0.04848566,0.09757379,0.003869569],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0005095899,0.0003700376,0.00232704,0.006907499,0.000155602,0.001606698,0.01510954,0.003961855,0.02282322,0.08610459,0.5219629,0.3381614],"study_design_scores_gemma":[0.00007666036,0.00006310987,0.002226153,0.002918876,0.00004258007,0.0002973517,0.002022759,0.01086969,0.01014704,0.05620028,0.9149173,0.0002181921],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001860067,0.0008171581,0.8635592,0.005234415,0.0009896122,0.008238288,0.03682927,0.05553166,0.0269403],"genre_scores_gemma":[0.0103612,0.001086997,0.8836172,0.001590451,0.0004366717,0.01533807,0.05619511,0.01593739,0.01543699],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0568324,"threshold_uncertainty_score":0.3005621,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2373113529841914,"score_gpt":0.4173617885935614,"score_spread":0.18005043560937,"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."}}