{"id":"W4309811661","doi":"10.1016/j.mex.2022.101935","title":"An automated method for developing search strategies for systematic review using Natural Language Processing (NLP)","year":2022,"lang":"en","type":"article","venue":"MethodsX","topic":"Species Distribution and Climate Change","field":"Environmental Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Forest Service","funders":"","keywords":"Computer science; Python (programming language); Data deduplication; Software; Systematic review; Data mining; Information retrieval; Artificial intelligence; Natural language processing; Machine learning; Database; Programming language","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.06149886,0.003865544,0.003451066,0.02458467,0.002045213,0.005289163,0.003063895,0.002094608,0.05530612],"category_scores_gemma":[0.1716217,0.002394859,0.007815334,0.01496348,0.00107236,0.00322501,0.005727672,0.002414057,0.01452068],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002853615,"about_ca_system_score_gemma":0.02431122,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004863013,"about_ca_topic_score_gemma":0.01465922,"domain_scores_codex":[0.9562515,0.02217485,0.014113,0.003194999,0.003841692,0.0004239147],"domain_scores_gemma":[0.8436036,0.1250282,0.009116468,0.00840636,0.01295604,0.000889258],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001119653,0.0002353896,0.002945359,0.08537533,0.003993331,0.0007176839,0.003289429,0.00456573,0.008547118,0.02354237,0.1420821,0.7235864],"study_design_scores_gemma":[0.004225174,0.0006833928,0.007177759,0.02496803,0.007404387,0.001501178,0.001414184,0.07537542,0.02296762,0.1195722,0.7335037,0.001206976],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002798059,0.003673451,0.8459417,0.002038663,0.0007888784,0.02539649,0.05549722,0.05750549,0.006360058],"genre_scores_gemma":[0.00433548,0.0006207389,0.9619142,0.0002851941,0.00006034632,0.02436082,0.006485352,0.001145393,0.0007923215],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9385011,"threshold_uncertainty_score":0.325241,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1108296255897259,"score_gpt":0.4701853592969272,"score_spread":0.3593557337072013,"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."}}