{"id":"W1595390009","doi":"","title":"Cybergenre: automatic identification of home pages on the web","year":2004,"lang":"en","type":"article","venue":"Journal of Web Engineering","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":40,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Web page; Home page; Computer science; World Wide Web; Classifier (UML); Information retrieval; Static web page; Web development; The Internet; Artificial intelligence","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":[],"consensus_categories":[],"category_scores_codex":[0.0009166237,0.00006649196,0.0001164013,0.0001343593,0.00003683809,0.0000456255,0.0004245586,0.00003166237,0.000007305181],"category_scores_gemma":[0.0001765336,0.00004607565,0.00007893566,0.0002991605,0.000009031046,0.0002121655,0.00003010009,0.0001845991,0.00001190241],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005262182,"about_ca_system_score_gemma":0.00007286236,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":2.999406e-7,"about_ca_topic_score_gemma":1.436199e-7,"domain_scores_codex":[0.9991562,0.00002519527,0.000394335,0.00006069526,0.0002593119,0.0001042282],"domain_scores_gemma":[0.9993175,0.0001034577,0.0002662639,0.0001848883,0.00008422934,0.00004369283],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000006743806,0.0001022107,0.0003001731,0.0001221658,0.0001095107,0.00003300303,0.001374622,0.3078537,0.4626285,0.2198997,0.0004764139,0.007093282],"study_design_scores_gemma":[0.0008904863,0.0002185661,0.02084105,0.0005574423,0.00003349299,0.000194929,0.0001156061,0.7145824,0.2571731,0.003853625,0.001230772,0.0003084721],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5975638,0.0002492008,0.3986022,0.002727661,0.0007237258,0.00004772744,9.687284e-7,0.00004096941,0.00004365709],"genre_scores_gemma":[0.9950804,0.00005709081,0.004741661,0.00003043736,0.00006999556,9.348668e-7,1.372431e-7,0.000004921963,0.0000144743],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4067287,"threshold_uncertainty_score":0.187891,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01742234240319396,"score_gpt":0.2319652182478575,"score_spread":0.2145428758446635,"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."}}