The application of Bayes' theorem in computer science dates back to the 1960s and continues to be heavily used in Na?veBayes classifiers in machine learning. In this report, we propose the use of a Na?ve Bayes-based classifier for automatedanalysis and data reduction of text-based log files generated by various computer systems and the services they provide. Theintended application of this technique is to automate the reduction of voluminous log files to a more manageable size and,with reasonable accuracy, retain log lines containing potential indicators of malicious cybersecurity activity or otherinfrequent interesting activity that should be examined further through other means.
展開▼
AI畢業(yè)論文
