{"id":"W2601473681","doi":"","title":"Enriching the legacy literature with OCR corrections and text-mined semantic metadata","year":2014,"lang":"en","type":"article","venue":"Research Explorer (The University of Manchester)","topic":"Species Distribution and Climate Change","field":"Environmental Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University; Open Text (Canada)","funders":"","keywords":"Computer science; Metadata; Information retrieval; Context (archaeology); Encyclopedia; Optical character recognition; Ranking (information retrieval); World Wide Web; Word (group theory); Natural language processing; Artificial intelligence","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0006857532,0.00008281043,0.00009651451,0.00004114603,0.0007079827,0.0001300104,0.0004415488,0.00003405587,0.001759852],"category_scores_gemma":[0.00006521353,0.0000494565,0.00003305047,0.0004355995,0.0007767144,0.0005703861,0.0004456328,0.0002824897,0.0001572076],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001071656,"about_ca_system_score_gemma":0.000007406742,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007321539,"about_ca_topic_score_gemma":0.0001308697,"domain_scores_codex":[0.9988387,0.0002420477,0.0000593067,0.0002130392,0.0004052386,0.0002416948],"domain_scores_gemma":[0.9992042,0.0002621104,0.00003889044,0.0003851238,0.00003452281,0.00007519053],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"observational","study_design_scores_codex":[0.0015047,0.001227124,0.2240309,0.0004892498,0.0005735476,0.0001551899,0.4730763,0.0003487953,0.02821752,0.03092053,0.1759455,0.0635106],"study_design_scores_gemma":[0.000962192,0.0002706858,0.4305858,0.000102532,0.00006401807,0.00003479095,0.1692041,0.001071492,0.0005374678,0.0002973122,0.3966326,0.0002370624],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9911688,0.00008035506,0.0008667537,0.004474685,0.00003064421,0.0002071395,0.0000096436,0.00001990881,0.003142044],"genre_scores_gemma":[0.9990979,0.0001940185,0.00005158123,0.00006804828,0.00001848999,0.000001199758,0.00001512629,0.000005685989,0.0005478944],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3038722,"threshold_uncertainty_score":0.9991527,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04346305473385294,"score_gpt":0.2529606173105499,"score_spread":0.2094975625766969,"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."}}