{"id":"W2095998031","doi":"10.1109/hicss.2007.98","title":"Automatic Web Page Categorization using Principal Component Analysis","year":2007,"lang":"en","type":"article","venue":"","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Information retrieval; Web page; World Wide Web; Categorization; Static web page; Web search engine; Web query classification; Web search query; Task (project management); Web navigation; Search engine; Artificial intelligence; 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":[],"consensus_categories":[],"category_scores_codex":[0.0003479547,0.0000854254,0.0001220947,0.0004549203,0.0001060912,0.0001360868,0.000499417,0.00005364512,0.00009327669],"category_scores_gemma":[0.00002129223,0.00007219263,0.00006294635,0.001649506,0.00003145655,0.0003514156,0.0001568724,0.00005343959,0.00003916827],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008198623,"about_ca_system_score_gemma":0.00002898522,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002880488,"about_ca_topic_score_gemma":0.00002850486,"domain_scores_codex":[0.9990577,0.00001753073,0.0002618794,0.0002335382,0.0002321013,0.0001972611],"domain_scores_gemma":[0.9992701,0.00004424276,0.0001098083,0.0004834757,0.00004760294,0.00004471057],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000002020173,0.0001824727,0.03862448,0.00001725919,0.0002322271,0.00001899413,0.0004753506,0.0009264566,0.05374307,0.8353376,0.0002033314,0.07023672],"study_design_scores_gemma":[0.00009377978,0.00001086658,0.02888267,0.000001464989,0.00003674364,0.00000224458,0.00009669636,0.9610277,0.008155668,0.001264437,0.0003122991,0.0001153793],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3016079,0.0000127804,0.6959523,0.0002463429,0.0000668464,0.00006541109,1.984998e-7,0.0007678144,0.001280436],"genre_scores_gemma":[0.9067155,0.000002779632,0.09304554,0.00005756228,0.00000919508,0.000002165131,0.000004139912,0.000002826672,0.0001602958],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9601013,"threshold_uncertainty_score":0.294393,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03073576962859396,"score_gpt":0.2876970805385032,"score_spread":0.2569613109099092,"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."}}