{"id":"W2036101886","doi":"10.5430/air.v2n1p44","title":"Exploiting web scraping in a collaborative filtering- based approach to web advertising","year":2012,"lang":"en","type":"article","venue":"Artificial Intelligence Research","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":132,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; World Wide Web; Web page; Scraper site; Copying; Web mining; Web development; Static web page; Web modeling; Web analytics; Web navigation; Information retrieval; Data Web; Web intelligence","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004107118,0.0008189813,0.001403587,0.003870361,0.00186177,0.003021067,0.001678249,0.00203931,0.001228508],"category_scores_gemma":[0.008080201,0.0006562243,0.001448448,0.003410746,0.001014588,0.002156233,0.001165856,0.001130376,0.0007827381],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009196628,"about_ca_system_score_gemma":0.001066492,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01105182,"about_ca_topic_score_gemma":0.01303954,"domain_scores_codex":[0.9963226,0.001187663,0.0003060323,0.0007165942,0.001249246,0.0002178116],"domain_scores_gemma":[0.9927911,0.003688662,0.0005472398,0.001390834,0.001341695,0.0002404893],"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":[0.0007397262,0.001446224,0.01036871,0.0004416389,0.0005957544,0.001224169,0.002428713,0.1158025,0.05623011,0.02530379,0.005564742,0.7798539],"study_design_scores_gemma":[0.00006089207,0.0002913091,0.004141542,0.00004074229,0.0002518052,0.00109633,0.0002595475,0.9397591,0.02396388,0.01537472,0.01459874,0.000161483],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04540999,0.000579598,0.9452137,0.0003739807,0.00007204533,0.0002872158,0.00006570897,0.002481796,0.005516069],"genre_scores_gemma":[0.4049489,0.0003696035,0.5882555,0.0002444978,0.0001336901,0.0001174673,0.0001821129,0.000137716,0.005610571],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01105182,"threshold_uncertainty_score":0.02197492,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2115775441977723,"score_gpt":0.410260957608213,"score_spread":0.1986834134104407,"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."}}