{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001532512,0.0001423247,0.0001991686,0.001150263,0.0002145467,0.0007319381,0.0008432502,0.00006660019,0.00001085177],"category_scores_gemma":[0.0001428673,0.0001209767,0.0001634508,0.0008497479,0.00002192024,0.01092675,0.0001352778,0.0001593526,0.00003913353],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001779537,"about_ca_system_score_gemma":0.0002128327,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001361819,"about_ca_topic_score_gemma":2.020567e-7,"domain_scores_codex":[0.9978149,0.00003250493,0.001334712,0.00005701628,0.0005002234,0.0002606738],"domain_scores_gemma":[0.9975631,0.00008602229,0.00089077,0.0002801372,0.001093207,0.00008674304],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0005282792,0.0001018538,0.00006834711,0.0007967409,0.0001649962,0.000001837906,0.001604688,0.004897124,0.0000908776,0.2452265,0.02249291,0.7240259],"study_design_scores_gemma":[0.005336939,0.0006327208,0.002074568,0.0002211432,0.00007020734,0.000009473571,0.000951676,0.3418508,0.005568952,0.002295446,0.6406677,0.0003204576],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001207691,0.00001313008,0.989193,0.002182947,0.0008114235,0.0007087304,0.000009132575,0.00005436149,0.005819539],"genre_scores_gemma":[0.3287656,0.0001035082,0.6502949,0.01887138,0.0001550623,0.00008801893,0.0001923622,0.00001429379,0.001514932],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7237054,"threshold_uncertainty_score":0.7921633,"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."}}