{"id":"W3080743972","doi":"10.11606/d.55.2020.tde-20082020-093906","title":"Interactive keyterm-based document clustering and visualization via neural language models","year":2020,"lang":"en","type":"dissertation","venue":"","topic":"Advanced Text Analysis Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Conselho Nacional de Desenvolvimento Científico e Tecnológico; Dalhousie University; Indiana Corn Marketing Council","keywords":"Cluster analysis; Visualization; Computer science; Natural language processing; Artificial neural network; Artificial intelligence","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006936283,0.0009462635,0.0005920544,0.001912792,0.0004735899,0.002668823,0.0009191135,0.0007518406,0.03171888],"category_scores_gemma":[0.004164279,0.0003505968,0.0008219026,0.001748687,0.0002279153,0.002607403,0.001373544,0.001248198,0.01148756],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005171529,"about_ca_system_score_gemma":0.0006853558,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004452305,"about_ca_topic_score_gemma":0.008483931,"domain_scores_codex":[0.9995632,0.0001022161,0.00003197209,0.0001194547,0.0001480681,0.00003507886],"domain_scores_gemma":[0.99867,0.0006460105,0.00005679855,0.0001891643,0.0003415315,0.00009653424],"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":[0.0006999358,0.0001749821,0.0009423905,0.0006143027,0.0001200783,0.0001880917,0.000458942,0.01827751,0.056385,0.006794925,0.08657652,0.8287673],"study_design_scores_gemma":[0.0001228583,0.0001270645,0.001951314,0.0001216782,0.00007646356,0.0002964693,0.0003613494,0.8238013,0.08255973,0.02313755,0.06733985,0.0001043426],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01535897,0.001220022,0.8998801,0.001071277,0.0004866452,0.0001417657,0.005605251,0.0697214,0.006514622],"genre_scores_gemma":[0.1405734,0.001283815,0.8262993,0.000204541,0.0002152702,0.0002440699,0.006681827,0.005326389,0.0191715],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03171888,"threshold_uncertainty_score":0.1061102,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009807791098549257,"score_gpt":0.3188874140990868,"score_spread":0.3090796230005375,"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."}}