{"id":"W2808777947","doi":"10.1145/3365376","title":"A Pure Visual Approach for Automatically Extracting and Aligning Structured Web Data","year":2019,"lang":"en","type":"article","venue":"ACM Transactions on Internet Technology","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Gestalt psychology; Information retrieval; Similarity (geometry); Web page; Task (project management); Information extraction; Data extraction; Object (grammar); Data mining; Artificial intelligence; Perception; Image (mathematics); World Wide Web","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007738493,0.001051755,0.0008699413,0.003797054,0.0005470847,0.00192126,0.001794977,0.001284562,0.002958526],"category_scores_gemma":[0.002604303,0.000705693,0.001234524,0.002492738,0.0008390784,0.002746535,0.001594994,0.001081238,0.002679684],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003773033,"about_ca_system_score_gemma":0.0009688927,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001996863,"about_ca_topic_score_gemma":0.002781177,"domain_scores_codex":[0.9989348,0.0001196241,0.00007795314,0.0004656109,0.0003351318,0.00006683174],"domain_scores_gemma":[0.9988813,0.0003024702,0.0001351385,0.0002595238,0.000339038,0.00008253798],"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":[0.0002104288,0.0001952029,0.001134358,0.0006518398,0.00007147642,0.0002133544,0.0002524794,0.002740514,0.1124836,0.003552024,0.007884198,0.8706106],"study_design_scores_gemma":[0.000148081,0.0006887511,0.01189457,0.0003021894,0.0003060845,0.003400032,0.0008933342,0.498207,0.3471235,0.02493569,0.1118473,0.0002535651],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.008394723,0.0004971866,0.9782833,0.0001273275,0.0000615397,0.000176238,0.000312666,0.01028263,0.001864358],"genre_scores_gemma":[0.05452394,0.0004971172,0.9404232,0.0002056229,0.00004513903,0.0001727069,0.001254915,0.0003419774,0.002535366],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003797054,"threshold_uncertainty_score":0.009897292,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02330713402028758,"score_gpt":0.2876187379659585,"score_spread":0.264311603945671,"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."}}