{"id":"W3204949062","doi":"10.18280/ria.350405","title":"Leveraging Pre-Trained Contextualized Word Embeddings to Enhance Sentiment Classification of Drug Reviews","year":2021,"lang":"en","type":"article","venue":"Revue d intelligence artificielle","topic":"Sentiment Analysis and Opinion Mining","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Sentiment analysis; Natural language processing; Leverage (statistics); Artificial intelligence; Word (group theory); Word embedding; Word2vec; Context (archaeology); Machine learning; Information retrieval; Embedding; Linguistics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"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":[0.0005840637,0.001466731,0.0007399636,0.001388729,0.0001990627,0.0006747353,0.0004783474,0.000673591,0.001333091],"category_scores_gemma":[0.002612977,0.0002265493,0.0007346684,0.0009920647,0.0002845405,0.001571697,0.0006368543,0.0009290127,0.00144678],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003149648,"about_ca_system_score_gemma":0.0004492264,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001812818,"about_ca_topic_score_gemma":0.003326707,"domain_scores_codex":[0.9995425,0.0001133111,0.00005516139,0.0001372224,0.00008946536,0.00006233656],"domain_scores_gemma":[0.9991533,0.000320085,0.0001271728,0.00007579439,0.0002843069,0.00003932319],"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":[0.0009817306,0.0006811143,0.01747546,0.000659698,0.0003518334,0.0004511099,0.0004026731,0.0488785,0.04712029,0.002375735,0.01402299,0.8665988],"study_design_scores_gemma":[0.00003996216,0.0003639598,0.006931921,0.00006635138,0.0001310078,0.000201096,0.0001917357,0.9707137,0.01148406,0.003635914,0.006196171,0.00004412542],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5721915,0.006976141,0.4034489,0.001034311,0.001173544,0.0003232872,0.002531259,0.005363149,0.006957786],"genre_scores_gemma":[0.9070929,0.001463152,0.08105767,0.0003838355,0.0003637635,0.0001742832,0.005099432,0.0001747808,0.004190214],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.001812818,"threshold_uncertainty_score":0.00445962,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05561749341251207,"score_gpt":0.3413495685495241,"score_spread":0.285732075137012,"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."}}