{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00520404,0.0001764393,0.0002576608,0.001165748,0.0003995555,0.0005499252,0.001293414,0.00007991378,0.00003163405],"category_scores_gemma":[0.001246729,0.0001768634,0.00006322897,0.006046857,0.0001215943,0.001115225,0.0006332644,0.0004917544,0.0004028331],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001897864,"about_ca_system_score_gemma":0.0003741456,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00020629,"about_ca_topic_score_gemma":0.0001171318,"domain_scores_codex":[0.9960811,0.0006150031,0.0005048226,0.0006446879,0.0008897783,0.00126462],"domain_scores_gemma":[0.9979173,0.0006006444,0.00005931632,0.0007178503,0.0003481738,0.0003567284],"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":[0.0001127466,0.001593866,0.007679521,0.0001334525,0.00006523605,0.00006004244,0.02757188,0.0217459,0.2460181,0.1228254,0.001186128,0.5710077],"study_design_scores_gemma":[0.00004111667,0.00009157789,0.0001491857,0.0001988698,0.00000319539,0.000002955248,0.007366637,0.8940472,0.09546468,0.0009153081,0.001381031,0.000338229],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1924749,0.0001937556,0.8006265,0.001074168,0.0001617072,0.0003460555,0.000009640809,0.0001235247,0.004989779],"genre_scores_gemma":[0.9295261,0.00001348185,0.07000553,0.0001225957,0.0001740355,0.00009555577,0.000006042329,0.00001584372,0.00004081642],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8723013,"threshold_uncertainty_score":0.721228,"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."}}