{"id":"W7008917920","doi":"","title":"Computing probabilities for common substrings in random strings","year":2006,"lang":"en","type":"dissertation","venue":"eScholarship@McGill (McGill)","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Substring; Set (abstract data type); Feature (linguistics); Term (time)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.004064764,0.0007338134,0.001328264,0.0097282,0.001175406,0.003916075,0.001839933,0.002450975,0.00443064],"category_scores_gemma":[0.05705109,0.0008639724,0.001827492,0.00546059,0.002186753,0.007167174,0.002818568,0.001976494,0.00204115],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00131834,"about_ca_system_score_gemma":0.001547213,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001978381,"about_ca_topic_score_gemma":0.002864222,"domain_scores_codex":[0.9925729,0.001623346,0.0009286319,0.002046348,0.002237126,0.0005916026],"domain_scores_gemma":[0.9565475,0.03465511,0.002133422,0.003707773,0.002312155,0.0006440032],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002118452,0.0004365102,0.05772868,0.001041764,0.0005257044,0.001768395,0.001444992,0.169558,0.01782623,0.1121678,0.01221561,0.623168],"study_design_scores_gemma":[0.0000763444,0.0002409556,0.01031237,0.0001674051,0.0001256821,0.001174789,0.0003708039,0.7486836,0.01062431,0.2244573,0.003670308,0.00009616384],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.260018,0.001358129,0.7293226,0.001046615,0.000229417,0.0002283424,0.002163455,0.002244488,0.00338901],"genre_scores_gemma":[0.7597697,0.001273394,0.2260132,0.0002450756,0.000435983,0.0003117102,0.007181064,0.0005326119,0.004237256],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0097282,"threshold_uncertainty_score":0.02149683,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01440460458459204,"score_gpt":0.2451875548206702,"score_spread":0.2307829502360782,"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."}}