{"id":"W4221142711","doi":"10.1145/3638243","title":"Building Domain-Specific Machine Learning Workflows: A Conceptual Framework for the State of the Practice","year":2023,"lang":"en","type":"article","venue":"ACM Transactions on Software Engineering and Methodology","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Polytechnique Montréal","funders":"Université de Montréal; McMaster University; McGill University","keywords":"Workflow; Executable; Computer science; Domain (mathematical analysis); Software engineering; Subject-matter expert; Key (lock); Automation; Software; Data science; Domain engineering; Artificial intelligence; Knowledge management; Software development; Expert system; Component-based software engineering; Engineering; Programming language; Database","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.04997301,0.001647102,0.001273538,0.008333741,0.003913264,0.02059988,0.009773659,0.006941212,0.003272598],"category_scores_gemma":[0.03847444,0.001941409,0.002913515,0.007397417,0.0292069,0.03453114,0.01015279,0.01046203,0.001916715],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006238338,"about_ca_system_score_gemma":0.01452076,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006136571,"about_ca_topic_score_gemma":0.002982551,"domain_scores_codex":[0.9737561,0.0159796,0.002450784,0.003114381,0.003553717,0.001145416],"domain_scores_gemma":[0.9560553,0.02235654,0.002485787,0.0125709,0.004617197,0.001914216],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00001084927,0.000047683,0.0002569986,0.0002081234,0.00001699416,0.00006709295,0.001665521,0.006444837,0.0002733579,0.9738462,0.000968823,0.01619347],"study_design_scores_gemma":[0.00001741669,0.00003977575,0.0001283252,0.0005367681,0.00001788844,0.0001232078,0.001419038,0.03252555,0.0008768041,0.9119448,0.0523174,0.00005304538],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001942036,0.000748458,0.9830357,0.00822595,0.00006881556,0.0001671139,0.00005900157,0.0004560782,0.005296955],"genre_scores_gemma":[0.04551832,0.001271089,0.950417,0.0008223749,0.0001078927,0.0004454142,0.0002150389,0.0001714515,0.001031373],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.950027,"threshold_uncertainty_score":0.2642857,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2062352037713582,"score_gpt":0.4093535291417279,"score_spread":0.2031183253703697,"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."}}