{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00022586,0.0001563235,0.0002375263,0.0004102667,0.00008553226,0.000141253,0.002141707,0.0002043538,0.00002964462],"category_scores_gemma":[0.0001117033,0.0001430294,0.00004517792,0.0004209226,0.00006660438,0.0004340589,0.0001595392,0.0003104963,0.00002099225],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000198636,"about_ca_system_score_gemma":0.00003644518,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001017412,"about_ca_topic_score_gemma":0.00001287198,"domain_scores_codex":[0.9986255,0.00002617837,0.0002444857,0.0007169703,0.0001243657,0.0002624836],"domain_scores_gemma":[0.9980382,0.000191855,0.00008887037,0.001595996,0.00004000823,0.00004507203],"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":[0.0000388838,0.0003131243,0.0007942408,0.0001568124,0.0005794202,0.00001294448,0.0007230082,0.0004373958,0.02146102,0.01301551,0.0008640796,0.9616036],"study_design_scores_gemma":[0.0005105025,0.0002605331,0.00004607376,0.00004580398,0.00005666485,0.0001350595,0.0004738737,0.989736,0.005048663,0.001804553,0.001651095,0.0002311307],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0857844,0.00002826813,0.9122235,0.00107939,0.0001178105,0.0001790183,0.00003996235,0.0004430046,0.0001046102],"genre_scores_gemma":[0.6873399,0.000004312679,0.3123406,0.00006253443,0.00001010374,0.00001879008,0.00002912454,0.000009154341,0.0001854679],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9892986,"threshold_uncertainty_score":0.5832571,"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."}}