{"id":"W4385573793","doi":"10.18653/v1/2022.emnlp-main.809","title":"Directions for NLP Practices Applied to Online Hate Speech Detection","year":2022,"lang":"en","type":"article","venue":"","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"European Social Fund; Fundação para a Ciência e a Tecnologia; Ministério da Ciência, Tecnologia e Ensino Superior","keywords":"Computer science; Conceptualization; Task (project management); Artificial intelligence; Generalization; Natural language processing; Situated; Quality (philosophy); Epistemology","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":[0.0004004646,0.0001083047,0.0001058107,0.0002087306,0.0007927401,0.0001538485,0.0004104909,0.00002934922,0.00008197681],"category_scores_gemma":[0.00009657598,0.00011125,0.00006165458,0.0008318687,0.000007201821,0.0002724128,0.0002442129,0.0001659608,0.00006258462],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001143071,"about_ca_system_score_gemma":0.00003939107,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002244935,"about_ca_topic_score_gemma":0.0004736978,"domain_scores_codex":[0.9988195,0.00004697598,0.0001741966,0.0004347622,0.0002631768,0.0002613924],"domain_scores_gemma":[0.9992133,0.0001011763,0.0001528842,0.0003650902,0.00006705694,0.0001004948],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00008514436,0.0002750149,0.00001148148,0.00001041093,0.00003163737,0.000004354662,0.0003396876,0.001515458,0.06987408,0.004655217,0.002442569,0.920755],"study_design_scores_gemma":[0.0005038405,0.0007507356,0.0003666269,0.000002471353,0.00001950594,0.0001140246,0.0005220076,0.03617115,0.1637212,0.002387505,0.7950692,0.000371799],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04713904,0.00002037187,0.9393777,0.002845682,0.00163708,0.0008749666,0.00001344891,0.0008477145,0.007244059],"genre_scores_gemma":[0.8215699,0.000007642369,0.1686828,0.001314355,0.0002656415,0.0007385169,0.00001279931,0.00002167708,0.007386665],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9203832,"threshold_uncertainty_score":0.6097196,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02623330174149566,"score_gpt":0.2818900551993356,"score_spread":0.2556567534578399,"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."}}