{"id":"W7116917707","doi":"10.1108/ajim-04-2025-0241","title":"LLM-based stemming for improved Gujarati information retrieval","year":2025,"lang":"en","type":"article","venue":"Aslib Journal of Information Management","topic":"Information Retrieval and Search Behavior","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; University of Saskatchewan","funders":"","keywords":"Gujarati; Relevance (law); Preprocessor; Ranking (information retrieval); Precision and recall; Recall; Language model; Focus (optics)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001344857,0.0007586451,0.0009729238,0.003555995,0.001088487,0.001917925,0.001162173,0.0006138594,0.002808247],"category_scores_gemma":[0.005395337,0.0002805467,0.001099748,0.003179396,0.0005319007,0.002032164,0.001493294,0.0006792917,0.003151662],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001397765,"about_ca_system_score_gemma":0.002734222,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01337843,"about_ca_topic_score_gemma":0.02021579,"domain_scores_codex":[0.998519,0.000437157,0.000198049,0.0003431014,0.0004043696,0.00009827495],"domain_scores_gemma":[0.9983913,0.0004213745,0.0001873119,0.0002308476,0.0007201295,0.00004888132],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0003649597,0.0001931781,0.009284035,0.002126073,0.000172554,0.0007083998,0.002298117,0.01382001,0.07527092,0.01164688,0.02158069,0.8625343],"study_design_scores_gemma":[0.0002180299,0.0009878059,0.0301683,0.0004900147,0.0006526626,0.003463853,0.003446182,0.623449,0.1223144,0.02600686,0.1883561,0.0004466974],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2591307,0.006124059,0.6746238,0.00249848,0.0004897904,0.001333357,0.00456829,0.02950284,0.02172864],"genre_scores_gemma":[0.4128307,0.001331041,0.5705245,0.0006435629,0.0001173956,0.0004319601,0.007255144,0.0008863584,0.005979375],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01337843,"threshold_uncertainty_score":0.02660114,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009732121916617402,"score_gpt":0.2624840641661548,"score_spread":0.2527519422495375,"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."}}