{"id":"W2598574795","doi":"10.29173/cais435","title":"Towards a Comparison of Structured and Unstructured Text Retrieval","year":2013,"lang":"en","type":"article","venue":"Proceedings of the Annual Conference of CAIS / Actes du congrès annuel de l ACSI","topic":"Information Retrieval and Search Behavior","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Information retrieval; Computer science; Adjacency list; Unstructured data; SGML; Text retrieval; Document Structure Description; Document retrieval; Natural language processing; World Wide Web; Data mining; XML; Big data; Algorithm","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003705244,0.0002260114,0.0005227239,0.0001860852,0.0001090465,0.001703042,0.002052907,0.0001614798,0.00003835571],"category_scores_gemma":[0.00441769,0.0001581953,0.0001228405,0.0006114125,0.0004837337,0.009152354,0.0008475596,0.000285567,0.000001983105],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003368369,"about_ca_system_score_gemma":0.0002447652,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001890735,"about_ca_topic_score_gemma":0.000002857242,"domain_scores_codex":[0.9979861,0.00002238161,0.0006785105,0.0002499101,0.0007201648,0.0003428686],"domain_scores_gemma":[0.9680382,0.00007190117,0.000828696,0.0002344925,0.03067617,0.0001505423],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0003877374,0.0002610078,0.2753512,0.001280746,0.0002030138,9.234603e-7,0.1645218,0.00001204748,0.2609484,0.1530104,0.00412787,0.1398948],"study_design_scores_gemma":[0.0009823245,0.0005841994,0.5578221,0.0001820391,0.00005004093,0.00003692355,0.002820057,0.007360013,0.4146904,0.01162103,0.003475046,0.0003758115],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9957159,0.00006916047,0.0003733479,0.001140635,0.0001275926,0.000426862,0.00005256301,0.00003544459,0.002058549],"genre_scores_gemma":[0.9957844,0.00002014551,0.003954267,0.00008063878,0.00002494054,0.00000690612,0.000001949044,0.000008393863,0.0001183353],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2824709,"threshold_uncertainty_score":0.9993333,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02263581327200936,"score_gpt":0.2646467163219727,"score_spread":0.2420109030499633,"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."}}