{"id":"W4402206897","doi":"10.1016/j.nlp.2024.100102","title":"Job description parsing with explainable transformer based ensemble models to extract the technical and non-technical skills","year":2024,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"Interpretability; Computer science; Margin (machine learning); Artificial intelligence; Machine learning; Transformer; Statistical model; Engineering","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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0007789218,0.0001925811,0.0001746356,0.0001758951,0.0004528939,0.001611788,0.0005038463,0.0001046594,0.000004143902],"category_scores_gemma":[0.00003219454,0.0001108476,0.00006359041,0.0004637141,0.00004958904,0.001686995,0.00004802849,0.001029409,0.000002315747],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001382816,"about_ca_system_score_gemma":0.0002452073,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001290605,"about_ca_topic_score_gemma":0.00003477754,"domain_scores_codex":[0.998373,0.00004820632,0.0002805291,0.0003821266,0.000494963,0.0004212162],"domain_scores_gemma":[0.9993805,0.00008840977,0.00005442308,0.0002192449,0.0001040167,0.0001534417],"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":[0.0000602988,0.00005614953,0.00003562614,0.0001650503,0.00001626643,0.0006323874,0.00671114,0.005842286,0.08360083,0.0006799905,0.0003176447,0.9018824],"study_design_scores_gemma":[0.0003345643,0.0001175842,0.0001522572,0.001169332,0.00003244985,0.004632707,0.0003251547,0.987662,0.003670765,0.001409413,0.0001949949,0.0002987745],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06002821,0.008850482,0.9276457,0.002395308,0.0001727277,0.000179363,5.175306e-7,0.0001897474,0.0005378876],"genre_scores_gemma":[0.8031062,0.00001202493,0.1961267,0.0004476308,0.0001466595,0.00001051842,4.948604e-7,0.00001679567,0.0001329393],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9818197,"threshold_uncertainty_score":0.9994246,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01353790584292541,"score_gpt":0.2620085156993179,"score_spread":0.2484706098563925,"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."}}