{"id":"W4393528389","doi":"10.5281/zenodo.7141166","title":"Kiswahili Monolingual Corpus","year":2022,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Global Maritime and Colonial Histories","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Natural language processing; Linguistics; Computer science; Artificial intelligence; Philosophy","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":[0.001042081,0.002240319,0.001017584,0.004332615,0.00248986,0.001729454,0.002050353,0.0018447,0.04772532],"category_scores_gemma":[0.003344038,0.0006053278,0.0006290319,0.006509103,0.0006507172,0.00184303,0.002363902,0.001724525,0.08407976],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002270932,"about_ca_system_score_gemma":0.004897619,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.05147628,"about_ca_topic_score_gemma":0.114807,"domain_scores_codex":[0.9989915,0.0002165839,0.0001473444,0.0002798963,0.0001727789,0.0001918909],"domain_scores_gemma":[0.9983622,0.0003396767,0.0001029091,0.0003405168,0.0006985558,0.0001560533],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0002262227,0.000124564,0.00153096,0.001544073,0.00004525119,0.0003146818,0.0005533648,0.0003171484,0.001299914,0.001258361,0.9762531,0.01653242],"study_design_scores_gemma":[0.0001189513,0.00003084506,0.009620696,0.0003927386,0.00004328952,0.0002664716,0.001281661,0.000589286,0.00148435,0.0008515332,0.9852628,0.00005744339],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.003309567,0.0003751979,0.000302765,0.0002167795,0.0001149713,0.00008523263,0.9901655,0.0005133093,0.004916677],"genre_scores_gemma":[0.001939333,0.00009555474,0.0006116927,0.00004948977,0.000009416068,0.0002448701,0.9946799,0.00006983671,0.002299943],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.05147628,"threshold_uncertainty_score":0.1596571,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03182316603825341,"score_gpt":0.2781406451941617,"score_spread":0.2463174791559082,"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."}}