{"id":"W2355952488","doi":"","title":"A Petroleum Subject Collection Strategy Based on the Word co-occurrence Model and the DOM","year":2008,"lang":"en","type":"article","venue":"Microcomputer applications","topic":"Advanced Computational Techniques and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Word (group theory); Categorization; Character (mathematics); Tree (set theory); Subject (documents); Artificial intelligence; Natural language processing; Tree structure; Information retrieval; Data structure; Linguistics; World Wide Web","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002151959,0.0001818474,0.0001451308,0.00008684202,0.001236884,0.0001470108,0.0009251618,0.00004600935,0.000002908189],"category_scores_gemma":[0.000001517283,0.0001171571,0.00008088899,0.0005926906,0.0002991744,0.0001327931,0.0001253681,0.0002425912,0.00002539681],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004701005,"about_ca_system_score_gemma":0.0001335759,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001001602,"about_ca_topic_score_gemma":0.000002352456,"domain_scores_codex":[0.9987729,0.00007473277,0.0002496335,0.0004729751,0.0002165026,0.0002133218],"domain_scores_gemma":[0.998524,0.0005380567,0.0001284431,0.000640884,0.0001066055,0.00006208169],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003352755,0.0002440322,0.00006844041,0.000009602616,0.00002128302,0.000001254267,0.0004076492,0.2122168,0.0003590172,0.7329677,0.01853796,0.0351327],"study_design_scores_gemma":[0.0004575282,0.00002307369,0.0004515866,0.00000636297,0.000005206293,0.00003499927,0.000005917932,0.925047,0.0003427343,0.03749555,0.03597593,0.0001541497],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001511898,0.00003722806,0.9883068,0.00770524,0.000006755221,0.001053841,0.00002778017,0.0002567031,0.001093792],"genre_scores_gemma":[0.677565,0.00006293445,0.3153961,0.003520619,0.00007103098,0.00318091,0.00003710348,0.00001266619,0.0001535966],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7128301,"threshold_uncertainty_score":0.951324,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02326029468827858,"score_gpt":0.2689521888825048,"score_spread":0.2456918941942262,"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."}}