{"id":"W4384664207","doi":"10.3758/s13428-023-02170-w","title":"LaDEP: A large database of English pseudo-compounds","year":2023,"lang":"en","type":"article","venue":"Behavior Research Methods","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Yorkville University; University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Database; Programming language; Information retrieval; Natural language processing; World Wide Web","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":[0.0008996624,0.001060673,0.0008736182,0.004707865,0.0007026929,0.00133006,0.00140979,0.001212478,0.01284493],"category_scores_gemma":[0.005451878,0.0004279928,0.0006558216,0.00482692,0.0004355515,0.002250275,0.001791387,0.000693947,0.009234064],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002652564,"about_ca_system_score_gemma":0.0007542691,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001665786,"about_ca_topic_score_gemma":0.002717278,"domain_scores_codex":[0.998693,0.0002601543,0.0004245871,0.0002604442,0.000290885,0.00007101758],"domain_scores_gemma":[0.9949692,0.002153333,0.0004443861,0.0009331706,0.001198516,0.0003013307],"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":[0.006023657,0.002440389,0.1103948,0.01406978,0.0008249609,0.01613652,0.004654428,0.002423047,0.07304693,0.003883352,0.1778705,0.5882316],"study_design_scores_gemma":[0.0006457425,0.001238782,0.3590641,0.0006588919,0.000452643,0.01410158,0.004449321,0.005004043,0.03469329,0.00232057,0.5770071,0.0003640305],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"dataset","genre_scores_codex":[0.5396248,0.003862648,0.01394765,0.0002731369,0.0001498087,0.001293663,0.4255691,0.002330936,0.01294826],"genre_scores_gemma":[0.2090409,0.001723966,0.03695416,0.0003349774,0.00008267158,0.002089068,0.742004,0.0006477303,0.007122633],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.01284493,"threshold_uncertainty_score":0.04297054,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2067183506943114,"score_gpt":0.5583345911829365,"score_spread":0.3516162404886251,"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."}}