{"id":"W207646885","doi":"10.1007/978-3-642-38457-8_11","title":"Selective Retrieval for Categorization of Semi-structured Web Resources","year":2013,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Categorization; Information retrieval; Ranking (information retrieval); Directory; Set (abstract data type); Process (computing); Task (project management); Entity linking; World Wide Web; Web page; Data mining; Artificial intelligence; Knowledge base","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001471736,0.0007610726,0.001390414,0.006092567,0.0009441382,0.001735446,0.001536569,0.0009067183,0.003452378],"category_scores_gemma":[0.003470605,0.0003126909,0.0009769775,0.004729412,0.0007899021,0.002981662,0.001899631,0.0007038682,0.003707876],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006855956,"about_ca_system_score_gemma":0.001430652,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002104992,"about_ca_topic_score_gemma":0.003977629,"domain_scores_codex":[0.9987971,0.000242549,0.0001427197,0.0002494775,0.0004263157,0.0001417295],"domain_scores_gemma":[0.9974883,0.001139062,0.0002069194,0.0005647012,0.0004710037,0.0001301034],"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":[0.0006273712,0.0003289922,0.00370119,0.001170834,0.0001730807,0.0003106954,0.000764538,0.002697813,0.07857682,0.007950078,0.0189944,0.8847041],"study_design_scores_gemma":[0.0002985239,0.001403149,0.03410978,0.0004623124,0.001076644,0.004570555,0.003839328,0.611559,0.1636632,0.1062834,0.07244467,0.0002895597],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1860995,0.006051462,0.7870535,0.0005602065,0.000208424,0.0008288503,0.004273124,0.007010245,0.007914694],"genre_scores_gemma":[0.4596707,0.002248462,0.5128947,0.0002996718,0.0003245962,0.0006554818,0.01366863,0.0004144417,0.009823339],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006092567,"threshold_uncertainty_score":0.01154941,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01253512869794304,"score_gpt":0.2330384927091037,"score_spread":0.2205033640111607,"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."}}