{"id":"W1535384510","doi":"10.16995/dscn.138","title":"Corpus Linguistics beyond Google: the WebCorp Linguist’s Search Engine","year":2009,"lang":"en","type":"article","venue":"Digital Studies / Le champ numérique","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Disk formatting; Computer science; World Wide Web; Information retrieval; Search engine; Web standards; Web search engine; Web page; Web search query","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":[0.004661338,0.0004101911,0.0007961015,0.004873935,0.002291593,0.006804808,0.001665344,0.002114953,0.04202892],"category_scores_gemma":[0.03278563,0.0006692822,0.0004083048,0.005350465,0.001551323,0.01501005,0.003101936,0.001945861,0.03055605],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002396934,"about_ca_system_score_gemma":0.00909312,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02233426,"about_ca_topic_score_gemma":0.05570256,"domain_scores_codex":[0.9970957,0.001047645,0.0002365282,0.0002505536,0.001193892,0.0001757193],"domain_scores_gemma":[0.9808476,0.007128284,0.0003696783,0.003462536,0.006215891,0.00197595],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001535958,0.00004058263,0.001205395,0.0004214482,0.00003204116,0.0001850268,0.0008545765,0.0001691867,0.001498512,0.04540158,0.7804013,0.1696367],"study_design_scores_gemma":[0.00002524285,0.00001373246,0.000944179,0.0003082261,0.00002047891,0.0001815697,0.0005244451,0.001641367,0.001684525,0.02281301,0.9718044,0.00003892115],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"software","genre_scores_codex":[0.02041766,0.02390561,0.0881675,0.3527772,0.008232077,0.00041522,0.03748738,0.03296776,0.4356296],"genre_scores_gemma":[0.2681929,0.02174596,0.1681684,0.03199978,0.008036609,0.0008625492,0.06022814,0.03042644,0.4103391],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.04202892,"threshold_uncertainty_score":0.1406008,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01986126175897783,"score_gpt":0.2839510981054043,"score_spread":0.2640898363464265,"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."}}