{"id":"W2908935617","doi":"10.1371/journal.pone.0212103","title":"Automating the search for a patent’s prior art with a full text similarity search","year":2019,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Intellectual Property and Patents","field":"Business, Management and Accounting","cited_by":68,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Berlin Center for Machine Learning; Institute for Information and Communications Technology Promotion; Banting and Best Diabetes Centre, University of Toronto; Bundesministerium für Bildung und Forschung","keywords":"Computer science; Novelty; Process (computing); Similarity (geometry); Information retrieval; Quality (philosophy); Invention; Domain (mathematical analysis); Symbol (formal); Patent application; Feature (linguistics); Nearest neighbor search; Artificial intelligence; Engineering","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0006541911,0.0001506015,0.0002026229,0.00009170402,0.0003540373,0.0003222863,0.0003175793,0.00005935543,0.001026568],"category_scores_gemma":[0.00009818437,0.00008823594,0.00005665405,0.0002959982,0.00005013246,0.0004157398,0.0001794925,0.000257971,0.003363033],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002445536,"about_ca_system_score_gemma":0.00003572791,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001520704,"about_ca_topic_score_gemma":0.00004721214,"domain_scores_codex":[0.9985826,0.00002320976,0.0001776634,0.0002752305,0.0005101492,0.0004311239],"domain_scores_gemma":[0.9991414,0.0001493203,0.00006153637,0.0002891182,0.0003447732,0.00001385964],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.01427064,0.01722567,0.5852048,0.03314702,0.005627155,0.00008985697,0.0109714,0.001652939,0.219731,0.01822548,0.03737798,0.05647602],"study_design_scores_gemma":[0.006753395,0.001156891,0.02844827,0.003036278,0.0007589433,0.00001099258,0.002206251,0.9109684,0.03082126,0.0006770331,0.0134481,0.001714167],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9781913,0.00002089244,0.0002402604,0.002003652,0.0000513456,0.001491522,0.000004534756,0.0001190582,0.01787739],"genre_scores_gemma":[0.9909723,0.000003308082,0.0005607976,0.001442965,0.0003792144,0.00006193467,0.00003088363,0.00003866131,0.006509966],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9093155,"threshold_uncertainty_score":0.9998866,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2542441161148024,"score_gpt":0.2361494102552961,"score_spread":0.01809470585950632,"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."}}