{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002635949,0.0001852653,0.0003774836,0.0004093259,0.0001319078,0.0003147954,0.00482469,0.00007587069,0.00003392817],"category_scores_gemma":[0.0001644005,0.0001501176,0.0001060385,0.0008659955,0.00001703539,0.000997876,0.001582874,0.0002507654,0.00005409565],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005193054,"about_ca_system_score_gemma":0.0002432363,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002473267,"about_ca_topic_score_gemma":0.000008648751,"domain_scores_codex":[0.9972265,0.0001976997,0.000772727,0.0007002352,0.0008455376,0.0002572666],"domain_scores_gemma":[0.9952167,0.0003483774,0.0004032256,0.003719409,0.0001483916,0.0001639238],"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":[0.0001912888,0.0007470266,0.001358422,0.00002488301,0.0008952962,0.00007443687,0.0005435708,0.02976087,0.08138376,0.0002326428,0.1957873,0.6890005],"study_design_scores_gemma":[0.0007380558,0.00005273505,0.0008227302,0.00006817859,0.0001190565,0.000002458113,0.00006875097,0.9925551,0.0008734606,0.0001161006,0.004421271,0.0001621265],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.005403598,0.0000953535,0.9890596,0.004123608,0.0006786087,0.0001209497,0.0002896982,0.00004960662,0.0001790298],"genre_scores_gemma":[0.3206645,0.0003397558,0.6700643,0.001533899,0.001050401,0.000005638697,0.005749079,0.00004904103,0.0005433843],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9627942,"threshold_uncertainty_score":0.8965554,"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."}}