{"id":"W2162076845","doi":"10.1145/1645953.1646270","title":"Data extraction from the web using wild card queries","year":2009,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Unary operation; Tuple; Information retrieval; Task (project management); Information extraction; Set (abstract data type); Web query classification; Simple (philosophy); Query language; Web search query; Search engine; Programming language","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":[0.00434625,0.001462434,0.002313375,0.009708743,0.001605912,0.005865606,0.002617325,0.00152237,0.00319064],"category_scores_gemma":[0.01764967,0.001054633,0.002070071,0.01111371,0.001479214,0.01323228,0.004428004,0.001988247,0.001919477],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009979369,"about_ca_system_score_gemma":0.001696368,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003692688,"about_ca_topic_score_gemma":0.004692691,"domain_scores_codex":[0.9940417,0.00137537,0.00113664,0.001130573,0.002079971,0.0002356198],"domain_scores_gemma":[0.9892561,0.005828906,0.0008148568,0.00280582,0.001080031,0.00021424],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0006723314,0.0005827517,0.01043986,0.003167015,0.0004301304,0.003293234,0.003990647,0.01989608,0.04370917,0.06028209,0.03939885,0.8141378],"study_design_scores_gemma":[0.0002398195,0.0003435774,0.006856044,0.0007200925,0.000442048,0.005840495,0.004283393,0.396136,0.09750126,0.2461682,0.2410682,0.0004009792],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01826384,0.002154573,0.9599087,0.001119873,0.00007713788,0.0006749692,0.005102391,0.009490691,0.003207813],"genre_scores_gemma":[0.09396028,0.00198024,0.8845831,0.0006299559,0.0001189805,0.0003928191,0.01597491,0.0006412106,0.00171847],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009708743,"threshold_uncertainty_score":0.02298546,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0866031512683388,"score_gpt":0.3185218247687534,"score_spread":0.2319186735004146,"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."}}