{"id":"W4296511188","doi":"10.4995/carma2022.2022.15076","title":"Text mining methods for innovation studies: limits and future perspectives","year":2022,"lang":"en","type":"article","venue":"","topic":"Expert finding and Q&A systems","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Word embedding; Word (group theory); Baseline (sea); Embedding; Information retrieval; Natural language processing; Ellipsis (linguistics); Turkish; Space (punctuation); Artificial intelligence; Linguistics","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":[0.03696646,0.001717437,0.002831219,0.006973298,0.001020816,0.01045379,0.004444327,0.003969079,0.007573299],"category_scores_gemma":[0.07713309,0.0008973992,0.001674911,0.009666012,0.004536364,0.01962908,0.003842798,0.003780738,0.003987289],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001772813,"about_ca_system_score_gemma":0.002614638,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002682945,"about_ca_topic_score_gemma":0.002064865,"domain_scores_codex":[0.9801086,0.01372832,0.0008823499,0.001643087,0.003359645,0.0002780563],"domain_scores_gemma":[0.7430893,0.2338536,0.003160009,0.0100715,0.008166341,0.001659288],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0003002563,0.0003547277,0.009846861,0.004014055,0.0004235514,0.0001898342,0.001248792,0.005705294,0.001150868,0.1765485,0.01781671,0.7824006],"study_design_scores_gemma":[0.0001188057,0.000302639,0.006112776,0.005525555,0.000139061,0.0005000499,0.003013796,0.07802662,0.001168132,0.6979735,0.2068891,0.0002299198],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"review","genre_gemma":"methods","genre_scores_codex":[0.01508346,0.4610945,0.3929539,0.09731258,0.002237913,0.0003965606,0.001192909,0.001141339,0.02858677],"genre_scores_gemma":[0.1759664,0.2445107,0.5426652,0.008033031,0.01232252,0.001741379,0.001397349,0.0005104688,0.01285304],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.03696646,"threshold_uncertainty_score":0.1954997,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08451033897803484,"score_gpt":0.4080580015405287,"score_spread":0.3235476625624938,"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."}}