{"id":"W4288019948","doi":"10.48550/arxiv.1912.01111","title":"Use of Artificial Intelligence to Analyse Risk in Legal Documents for a\\n Better Decision Support","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"PricewaterhouseCoopers (Canada)","funders":"","keywords":"Computer science; Paragraph; Word embedding; Artificial intelligence; Machine learning; Context (archaeology); Scalability; Support vector machine; Document classification; Natural language processing; Information retrieval; Data science; Knowledge management; Embedding; World Wide Web; Database","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":[0.002617695,0.0006956755,0.0004565485,0.009372828,0.0007209401,0.003463514,0.0007285026,0.0008339874,0.002548962],"category_scores_gemma":[0.01567333,0.0002147329,0.0006142775,0.003748454,0.0007527514,0.003566347,0.0008115213,0.001084284,0.0009026694],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001401185,"about_ca_system_score_gemma":0.001376116,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004462013,"about_ca_topic_score_gemma":0.005527461,"domain_scores_codex":[0.9970175,0.001247058,0.0003630337,0.0003587649,0.0009245364,0.00008920907],"domain_scores_gemma":[0.9899983,0.006271169,0.001423671,0.0008958019,0.001242281,0.0001688662],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001547396,0.0002935406,0.03058234,0.0006197931,0.0002358791,0.0006333201,0.0008872799,0.05944879,0.01025957,0.04035144,0.01464804,0.8418853],"study_design_scores_gemma":[0.00002902498,0.000138131,0.01520759,0.0003102595,0.0001134508,0.000529422,0.0009670368,0.7744567,0.01944288,0.1464835,0.04218515,0.0001368917],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1755525,0.004590909,0.7734694,0.006125395,0.0003411558,0.0005929233,0.004362239,0.005333837,0.0296316],"genre_scores_gemma":[0.563117,0.001356937,0.4272701,0.0003243831,0.0001565718,0.0001456315,0.003674418,0.0001391471,0.00381599],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009372828,"threshold_uncertainty_score":0.01384389,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1927989205076086,"score_gpt":0.3117018278638882,"score_spread":0.1189029073562796,"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."}}