{"id":"W7127444902","doi":"10.1109/ccece64018.2025.11364472","title":"Advanced Feature Engineering for Twitter Bot Detection: Utilizing Metadata, NLP, and Transformers","year":2025,"lang":"","type":"article","venue":"","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"New York Institute of Technology","funders":"","keywords":"Transformer; Feature extraction; Feature engineering; Feature (linguistics); Natural language; Random forest; Deep learning; Feature learning","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003903773,0.0003449538,0.0003322376,0.0003006006,0.0004403447,0.0008250312,0.0003382753,0.0002341582,0.00001318924],"category_scores_gemma":[0.0000723038,0.000343963,0.0001748004,0.0008135157,0.00004189924,0.001980592,0.0000894824,0.0003771139,0.000002275391],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008486876,"about_ca_system_score_gemma":0.00006751004,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004103057,"about_ca_topic_score_gemma":0.0000828819,"domain_scores_codex":[0.9981741,0.00002890297,0.0003150593,0.0007871976,0.0001909008,0.0005038168],"domain_scores_gemma":[0.9991275,0.0002105973,0.00006300913,0.000368475,0.0001038869,0.0001265142],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001654294,0.00003543521,0.00003789212,0.0005398624,0.0002254877,0.000002885233,0.0008628438,0.0007219258,0.04154708,0.006236515,0.0006365465,0.9489881],"study_design_scores_gemma":[0.001991884,0.0004177865,0.0011631,0.0002769346,0.0002153261,0.00003830365,0.0003374417,0.5639934,0.2032472,0.0007409262,0.226878,0.0006997122],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00980397,0.002335947,0.9795305,0.002564852,0.00390876,0.0006685161,0.000006036798,0.0002092607,0.0009721229],"genre_scores_gemma":[0.9551674,0.000212336,0.0379831,0.0007626452,0.000169943,0.00008467883,0.000002508289,0.00002139066,0.005596068],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9482884,"threshold_uncertainty_score":0.9999012,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01242306746153325,"score_gpt":0.2450554857543261,"score_spread":0.2326324182927929,"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."}}