{"id":"W2074118501","doi":"10.3115/1641976.1641987","title":"A measure of aggregate syntactic distance","year":2006,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Oulun Yliopisto; Rijksuniversiteit Groningen; Université de Moncton","keywords":"Trigram; Syntax; Computer science; Natural language processing; Measure (data warehouse); Aggregate (composite); Artificial intelligence; Permutation (music); Syntactic structure","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"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":[0.000109256,0.0000633949,0.00008765206,0.00004336917,0.00002591039,0.00004322153,0.0005205898,0.00003016732,0.000006988038],"category_scores_gemma":[0.00003099782,0.00004804498,0.00002959922,0.000270831,0.00002559592,0.0003135805,0.00007669171,0.00006152911,0.000005067791],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001817131,"about_ca_system_score_gemma":0.000023827,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001412296,"about_ca_topic_score_gemma":0.00003329644,"domain_scores_codex":[0.9994031,0.00001618523,0.0001229379,0.0001557342,0.0001878405,0.0001141538],"domain_scores_gemma":[0.9994868,0.00002840034,0.0000799583,0.0003092463,0.00007850044,0.00001707469],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000003717595,0.00004439527,0.0004774719,0.0000386575,0.000004011476,0.0000184737,0.000050556,0.00000184791,0.02218556,0.9410521,0.001442799,0.03468043],"study_design_scores_gemma":[0.0001496692,0.00004111268,0.0004030114,0.0001323681,0.000005326432,0.0000307074,0.000004520044,0.004274385,0.6982051,0.2949965,0.001526422,0.0002308535],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0014925,0.002466732,0.9879037,0.000406753,0.00004169055,0.00005177134,5.022505e-7,0.0004522849,0.007184055],"genre_scores_gemma":[0.6335763,0.000001369301,0.365937,0.00004044806,0.00001006243,0.000002105209,2.315977e-7,0.000002403388,0.0004300564],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6760195,"threshold_uncertainty_score":0.1959217,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007379287816726015,"score_gpt":0.2348163491163288,"score_spread":0.2274370612996028,"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."}}