{"id":"W2996771664","doi":"10.1109/saner48275.2020.9054792","title":"Cross-Dataset Design Discussion Mining","year":2020,"lang":"en","type":"preprint","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Victoria","funders":"","keywords":"Computer science; Classifier (UML); Transfer of learning; Artificial intelligence; Machine learning; Documentation; Software; Replicate; Data mining","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":[0.009887331,0.001286797,0.0007414073,0.01127084,0.002332848,0.003417261,0.003113309,0.002401001,0.01143179],"category_scores_gemma":[0.04082252,0.0005274744,0.001558831,0.008328183,0.0007764983,0.002873199,0.003918207,0.002155685,0.006921145],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001860923,"about_ca_system_score_gemma":0.002659931,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003515407,"about_ca_topic_score_gemma":0.0070151,"domain_scores_codex":[0.9847087,0.005093238,0.001909381,0.003558557,0.00403013,0.0007000549],"domain_scores_gemma":[0.9562793,0.01769404,0.003931799,0.01290257,0.008007907,0.001184518],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00124155,0.002307107,0.1011183,0.006388477,0.0009463764,0.0009844906,0.003808409,0.007040284,0.01387715,0.01564094,0.3336844,0.5129625],"study_design_scores_gemma":[0.0003548983,0.0004626002,0.08795677,0.0007677053,0.0004051254,0.001008522,0.003438568,0.02558269,0.02390143,0.01282956,0.8431137,0.0001784842],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.2972983,0.007065965,0.1237271,0.003233862,0.00110716,0.006894808,0.4575071,0.01193056,0.09123517],"genre_scores_gemma":[0.259554,0.0008047222,0.1317608,0.001237101,0.0002555971,0.007140467,0.5757498,0.001090987,0.02240647],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.01143179,"threshold_uncertainty_score":0.05228984,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08501070002276262,"score_gpt":0.350103306526252,"score_spread":0.2650926065034894,"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."}}