{"id":"W3168278108","doi":"10.1017/pan.2021.37","title":"Cross-Domain Topic Classification for Political Texts","year":2021,"lang":"en","type":"article","venue":"Political Analysis","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":47,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"H2020 European Research Council; University of Essex; Università Bocconi; Eidgenössische Technische Hochschule Zürich; Wissenschaftszentrum Berlin für Sozialforschung; York University; London School of Economics and Political Science","keywords":"Computer science; Classifier (UML); Domain (mathematical analysis); Artificial intelligence; Natural language processing; Labeled data; Machine learning; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0007856417,0.0001055195,0.0003253566,0.0001876098,0.0005060387,0.0002035653,0.0001762034,0.0001120647,0.001600706],"category_scores_gemma":[0.002016568,0.00010191,0.0006446657,0.001447743,0.0003368338,0.00009907457,0.00004120969,0.00008955732,0.000051118],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002061624,"about_ca_system_score_gemma":0.0002901785,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009328364,"about_ca_topic_score_gemma":0.001083222,"domain_scores_codex":[0.9976563,0.0004228544,0.0003549136,0.0003862333,0.0004512079,0.000728459],"domain_scores_gemma":[0.9973983,0.001274734,0.00006703155,0.0002221348,0.0005577386,0.000480123],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000003705971,0.00008464647,0.03590837,0.000006173726,0.0003500464,0.000003177129,0.0001595834,0.0000154646,0.00003819566,0.9610949,0.00004632563,0.002289372],"study_design_scores_gemma":[0.0002150421,0.00001210954,0.180781,0.000002922412,0.001080018,8.4032e-7,0.001651643,0.00556631,0.0001711143,0.7706351,0.03967771,0.0002062131],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2049179,0.0001622702,0.5352697,0.0642454,0.0002497449,0.0002179305,0.00009016116,0.0001465781,0.1947003],"genre_scores_gemma":[0.974386,0.000002272496,0.01544651,0.001074984,0.0006297972,0.00002857651,0.0001058904,0.000006734825,0.008319268],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.769468,"threshold_uncertainty_score":0.999312,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08141688647427724,"score_gpt":0.4642052680106769,"score_spread":0.3827883815363996,"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."}}