{"id":"W3031973162","doi":"","title":"NLP Scholar: A Dataset for Examining the State of NLP Research.","year":2020,"lang":"en","type":"article","venue":"NPARC","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Metadata; Artificial intelligence; Information retrieval; Natural language processing; Citation; Focus (optics); Named entity; World Wide Web","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch","bibliometrics"],"consensus_categories":[],"category_scores_codex":[0.003287061,0.001308647,0.001202925,0.03003206,0.002014078,0.004114831,0.002352022,0.002488571,0.0144493],"category_scores_gemma":[0.02365172,0.0004757541,0.001218584,0.03896389,0.0007311793,0.00355666,0.003436443,0.001678092,0.01820155],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002924765,"about_ca_system_score_gemma":0.00523096,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01845892,"about_ca_topic_score_gemma":0.04191108,"domain_scores_codex":[0.9935185,0.001040958,0.001644323,0.0009558341,0.002420264,0.0004200538],"domain_scores_gemma":[0.979138,0.008978751,0.003526271,0.00246099,0.004452509,0.001443541],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001894471,0.0001127193,0.007987071,0.005166865,0.0001200006,0.0004504282,0.0006480373,0.0006633102,0.001817368,0.008138498,0.9508665,0.02383985],"study_design_scores_gemma":[0.0001119409,0.00002804363,0.00945285,0.0003720282,0.00004040232,0.0002693106,0.0005183585,0.0009601485,0.001134818,0.002888758,0.9841843,0.00003908457],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.003212762,0.001416893,0.0008658606,0.0007096549,0.0001499843,0.0001034308,0.9853247,0.0009615825,0.007255074],"genre_scores_gemma":[0.003089621,0.0006146579,0.002973101,0.0001447227,0.00006232407,0.0001824687,0.9914403,0.0001237185,0.001368961],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.9967129,"threshold_uncertainty_score":0.0483377,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1219329346152474,"score_gpt":0.3756270129000298,"score_spread":0.2536940782847824,"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."}}