{"id":"W4392374041","doi":"10.20944/preprints202402.1634.v1","title":"Enhancing Legal Sentiment Analysis: A CNN-LSTM Document-Level Model","year":2024,"lang":"en","type":"preprint","venue":"Preprints.org","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Athabasca University","funders":"","keywords":"Sentiment analysis; Computer science; Natural language processing; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.003683742,0.0005990898,0.000872603,0.0006099337,0.0006522341,0.0004374748,0.001747569,0.0007081331,0.00518115],"category_scores_gemma":[0.0005169916,0.000661303,0.001047603,0.001079288,0.0005572074,0.000309268,0.004887436,0.001764184,0.009406234],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001166455,"about_ca_system_score_gemma":0.001393714,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02370915,"about_ca_topic_score_gemma":0.01672183,"domain_scores_codex":[0.9935356,0.0004790437,0.001273147,0.002038991,0.001612551,0.001060634],"domain_scores_gemma":[0.9968728,0.0001669019,0.0004720588,0.001687218,0.000370813,0.0004302053],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001405467,0.0008167168,0.06938712,0.0007202263,0.01023839,0.0002589159,0.1476028,0.3045924,0.01712295,0.4445149,0.0007732132,0.003831786],"study_design_scores_gemma":[0.0002079703,0.00003212985,0.003835194,0.001228115,0.007545974,0.000003446472,0.01553573,0.07476639,0.2692509,0.588329,0.03543393,0.003831323],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8230894,0.0003629016,0.01111936,0.003196748,0.00279809,0.001433485,0.00007547204,0.0008469063,0.1570777],"genre_scores_gemma":[0.9596047,0.0001682895,0.00127081,0.0002259243,0.0007228356,0.0003312679,0.00003099639,0.00006860937,0.03757651],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2521279,"threshold_uncertainty_score":0.9995838,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2204475637884853,"score_gpt":0.4413228074950138,"score_spread":0.2208752437065286,"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."}}