{"id":"W4312443713","doi":"10.1109/re54965.2022.00011","title":"Automated Question Answering for Improved Understanding of Compliance Requirements: A Multi-Document Study","year":2022,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Jargon; Computer science; Compliance (psychology); Question answering; Natural language; Requirements engineering; Software requirements; Subject (documents); Information retrieval; Software; World Wide Web; Artificial intelligence; Software development; Software design; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01220321,0.0009959381,0.0009920289,0.004796665,0.001472076,0.002866386,0.001755817,0.002174887,0.002997954],"category_scores_gemma":[0.04936987,0.000448083,0.001075666,0.002785879,0.0009301672,0.005841664,0.00284168,0.002092442,0.001377921],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001529364,"about_ca_system_score_gemma":0.001730591,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005737309,"about_ca_topic_score_gemma":0.004811529,"domain_scores_codex":[0.9815788,0.0121538,0.001125032,0.002410653,0.002370424,0.0003611839],"domain_scores_gemma":[0.8583996,0.123759,0.004141656,0.00428017,0.008365941,0.001053633],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001249279,0.002834874,0.03899856,0.003488651,0.0004350696,0.001308237,0.01709894,0.02567684,0.05279904,0.007718139,0.01458113,0.8338112],"study_design_scores_gemma":[0.0004881958,0.002040392,0.07371753,0.0007393383,0.0005700717,0.003060953,0.01483421,0.749949,0.06301175,0.01411399,0.07714566,0.0003289397],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6881469,0.004972413,0.2895168,0.002397597,0.0001223002,0.001414032,0.002920229,0.003976409,0.006533188],"genre_scores_gemma":[0.6974937,0.001169002,0.2881874,0.000583639,0.0001284099,0.0006804222,0.008430876,0.0004647485,0.002861924],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01220321,"threshold_uncertainty_score":0.06453753,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1345397905093531,"score_gpt":0.380177912239094,"score_spread":0.2456381217297409,"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."}}