{"id":"W4407368063","doi":"10.2139/ssrn.5071032","title":"Use of text mining and natural language processing techniques in analyzing sustainability reports: A systematic literature review and assessment","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Corporate Social Responsibility Reporting","field":"Business, Management and Accounting","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Ottawa","funders":"","keywords":"Sustainability; Natural (archaeology); Systematic review; Computer science; Natural language processing; Data science; Political science; History; MEDLINE; Archaeology; Biology","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":[0.04849714,0.001427503,0.005244043,0.03817173,0.001109522,0.003893746,0.002261751,0.001811379,0.002020371],"category_scores_gemma":[0.1278732,0.001040997,0.007021148,0.02429687,0.002314113,0.005435807,0.002602844,0.001446182,0.0003445166],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00256065,"about_ca_system_score_gemma":0.01951477,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00547797,"about_ca_topic_score_gemma":0.01750553,"domain_scores_codex":[0.9647888,0.01092297,0.01306086,0.003134385,0.007648708,0.0004442241],"domain_scores_gemma":[0.7127445,0.2370945,0.02867359,0.004199301,0.01632938,0.0009587626],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"systematic_review","study_design_scores_codex":[0.0003245933,0.0002085134,0.01621591,0.6247777,0.008683021,0.000345251,0.001957697,0.0004938084,0.001692457,0.0007398281,0.00193296,0.3426283],"study_design_scores_gemma":[0.0002762288,0.001245046,0.05064272,0.7813557,0.08774167,0.001763166,0.005615766,0.00149812,0.004743769,0.001932593,0.06289449,0.0002906924],"study_design_candidate":"systematic_review","study_design_consensus":"systematic_review","genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.01963836,0.9683619,0.005030533,0.001169933,0.000188821,0.002079369,0.002459332,0.00006774387,0.001003989],"genre_scores_gemma":[0.09541842,0.8717489,0.02626716,0.001348152,0.0001995951,0.002469213,0.002160853,0.00004994463,0.0003376474],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.04849714,"threshold_uncertainty_score":0.2564805,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01153903176317778,"score_gpt":0.3095151938633435,"score_spread":0.2979761621001657,"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."}}