{"id":"W4324378562","doi":"10.1186/s40537-023-00710-x","title":"A semi-supervised short text sentiment classification method based on improved Bert model from unlabelled data","year":2023,"lang":"en","type":"article","venue":"Journal Of Big Data","topic":"Sentiment Analysis and Opinion Mining","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Computer science; Bottleneck; Sentiment analysis; Artificial intelligence; Semi-supervised learning; Machine learning; Big data; Supervised learning; Language model; Function (biology); Natural language processing; Data mining; Artificial neural network","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":[0.0009634281,0.001057154,0.0009585054,0.001172964,0.0006058888,0.0008097489,0.001787452,0.0009995655,0.002227114],"category_scores_gemma":[0.001627171,0.0003887465,0.0009008281,0.0007097026,0.0003929238,0.001583818,0.0007188633,0.001035018,0.001178084],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000794247,"about_ca_system_score_gemma":0.0008863439,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00639329,"about_ca_topic_score_gemma":0.007446339,"domain_scores_codex":[0.9994112,0.00009987212,0.00004396329,0.0001726903,0.0001902004,0.00008214654],"domain_scores_gemma":[0.9992041,0.0001792415,0.00006884312,0.00007104863,0.0004241559,0.00005256534],"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":[0.0008885717,0.0005879161,0.008885298,0.0002020723,0.0001521733,0.0003386301,0.000261733,0.2325515,0.03202963,0.003731552,0.01346283,0.706908],"study_design_scores_gemma":[0.000005994061,0.00002642294,0.0003550702,0.000003307843,0.000007490577,0.0000140565,0.000009585263,0.9974112,0.001453102,0.0004174055,0.0002907176,0.000005699204],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1452592,0.0006761482,0.8440723,0.0006695605,0.0003192094,0.0002584615,0.0005136308,0.003615922,0.004615513],"genre_scores_gemma":[0.8554309,0.0003595477,0.1295015,0.0004128112,0.000295052,0.0003256378,0.00248665,0.0001778676,0.01101],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00639329,"threshold_uncertainty_score":0.01271212,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2707046506402733,"score_gpt":0.3750577178151738,"score_spread":0.1043530671749005,"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."}}