{"id":"W1494273468","doi":"10.1007/978-3-540-24840-8_5","title":"Term-Based Clustering and Summarization of Web Page Collections","year":2004,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":false,"ca_institutions":"Dalhousie University","funders":"","keywords":"Automatic summarization; Computer science; Cluster analysis; Information retrieval; Web page; World Wide Web; Web intelligence; Task (project management); Web modeling; 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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004365754,0.0002820082,0.0003846624,0.001140857,0.0002753663,0.0004028165,0.001242488,0.0001731147,0.00001261212],"category_scores_gemma":[0.00006463222,0.0002693231,0.00008057435,0.001077445,0.0004591113,0.0003648365,0.0006730312,0.0003072463,0.000002657253],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001494329,"about_ca_system_score_gemma":0.0006865841,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004512096,"about_ca_topic_score_gemma":0.0001924417,"domain_scores_codex":[0.9979002,0.00002302987,0.0003910164,0.000893255,0.0004977742,0.0002947011],"domain_scores_gemma":[0.9984469,0.0002251633,0.0002502747,0.0008357617,0.0001453068,0.00009661455],"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":[0.00001341576,0.0001148407,0.001041082,0.000335728,0.00006477642,0.0001014929,0.001338232,0.3691297,0.005494097,0.01458178,0.00005057679,0.6077342],"study_design_scores_gemma":[0.0003110649,0.0001120651,0.0001270811,0.0005194627,0.00001809841,0.00001529516,1.372633e-7,0.9848818,0.00128085,0.01224959,0.000140927,0.0003436384],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0002029377,0.0001210006,0.9973555,0.0002689094,0.0003718062,0.0001323673,0.0000158745,0.00007745039,0.001454132],"genre_scores_gemma":[0.5984342,0.00004827886,0.4006336,0.0003666832,0.0001304217,0.000004888792,0.00001807426,0.00002190988,0.0003420455],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.615752,"threshold_uncertainty_score":0.9999759,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01384464157001871,"score_gpt":0.2280352087547077,"score_spread":0.214190567184689,"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."}}