{"id":"W4391814456","doi":"10.1017/asb.2024.4","title":"Telematics combined actuarial neural networks for cross-sectional and longitudinal claim count data","year":2024,"lang":"en","type":"article","venue":"Astin Bulletin","topic":"Insurance, Mortality, Demography, Risk Management","field":"Social Sciences","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Negative binomial distribution; Count data; Artificial neural network; Telematics; Computer science; Interpretability; Generalized linear model; Component (thermodynamics); Perceptron; Poisson distribution; Poisson regression; Data mining; Artificial intelligence; Machine learning; Statistics; Mathematics; Telecommunications","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002129137,0.0001835047,0.0002082717,0.00008410757,0.0007824086,0.000902946,0.0005244867,0.0001349959,0.0004559045],"category_scores_gemma":[0.0003191058,0.0001797304,0.00008598089,0.0002407259,0.0005072736,0.0001547615,0.0002670079,0.0002415458,0.00003552781],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005579297,"about_ca_system_score_gemma":0.00006306793,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006795662,"about_ca_topic_score_gemma":0.0006662033,"domain_scores_codex":[0.9979194,0.0001208518,0.0003897742,0.0005802889,0.000511509,0.0004781516],"domain_scores_gemma":[0.9986516,0.0006639477,0.00009036166,0.0003665711,0.0001186954,0.0001087941],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0002862163,0.0001609056,0.6783788,0.0003292948,0.0003125359,0.00005007604,0.0008986132,0.001456329,0.000002742492,0.07966086,0.2296165,0.008847103],"study_design_scores_gemma":[0.0008059964,0.0001269566,0.4335096,0.00005190395,0.00009869008,0.000005997116,0.0002396349,0.06074585,7.147507e-7,0.002142098,0.5019004,0.0003721413],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8884462,0.002514344,0.05758787,0.01428845,0.01800718,0.004247262,0.0008585336,0.00107872,0.01297143],"genre_scores_gemma":[0.9950656,0.00010125,0.001583138,0.0002058502,0.001900854,0.00006449713,0.0001785782,0.00002736001,0.0008728647],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2722839,"threshold_uncertainty_score":0.8707129,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06105587676755898,"score_gpt":0.3567739675076773,"score_spread":0.2957180907401183,"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."}}