{"id":"W3090026837","doi":"10.1007/s10044-020-00917-1","title":"Efficient integration of generative topic models into discriminative classifiers using robust probabilistic kernels","year":2020,"lang":"en","type":"article","venue":"Pattern Analysis and Applications","topic":"Image Retrieval and Classification Techniques","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université TÉLUQ; Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Discriminative model; Artificial intelligence; Dirichlet distribution; Pattern recognition (psychology); Decision boundary; Latent Dirichlet allocation; Computer science; Generative model; Support vector machine; Mathematics; Classifier (UML); Prior probability; Machine learning; Bayesian probability; Topic model; Generative grammar","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001798669,0.0008727878,0.001759033,0.001150838,0.0005150363,0.001663243,0.001767356,0.001272844,0.002075198],"category_scores_gemma":[0.005571269,0.0007422617,0.001477614,0.001546154,0.000409879,0.003206978,0.002216164,0.00196892,0.002671303],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007077172,"about_ca_system_score_gemma":0.001036275,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003492255,"about_ca_topic_score_gemma":0.006673226,"domain_scores_codex":[0.9987988,0.000327386,0.00007020674,0.0003100646,0.0003368214,0.0001565984],"domain_scores_gemma":[0.9977616,0.001048337,0.0001397461,0.0004698177,0.000476765,0.0001037317],"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":[0.0004090529,0.000282944,0.001970968,0.0001533574,0.0002218388,0.00008822714,0.0001816514,0.1522449,0.02626092,0.01638819,0.008154329,0.7936436],"study_design_scores_gemma":[0.000007647222,0.00002334406,0.0003005986,0.000004301284,0.00002919383,0.00003898062,0.00001316839,0.9901298,0.002955164,0.005650457,0.0008376752,0.000009770494],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.012718,0.0003417957,0.9843568,0.0001133625,0.00004408028,0.00002484859,0.00008254884,0.001787289,0.0005312652],"genre_scores_gemma":[0.5369628,0.000719303,0.452911,0.000227137,0.0002130833,0.0001667917,0.002075957,0.0009196202,0.005804326],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003492255,"threshold_uncertainty_score":0.009512424,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07222202404860935,"score_gpt":0.2936769322880273,"score_spread":0.221454908239418,"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."}}