A Survey on Anomaly Detection Methods for System Log Data
Journal: International Journal of Science and Research (IJSR) (Vol.8, No. 7)Publication Date: 2019-07-05
Authors : Devika Ajith;
Page : 323-325
Keywords : Deep learning; machine learning; anomaly detection; deeplog; artificial intelligence; natural language processing;
Abstract
System logs are often a collection of unrelated print statements which records certain events that occur while the system is running. Log file analysis often can be crucial for finding system faults which can otherwise be quite difficult to detect. Traditional log analysis involves analyzing line by line until a discrepancy is spotted. This process is tedious, time consuming and is prone to human errors. With the advent of machine learning, several new methods have been devised which can make anomaly detection much easier. Further, in the past decade, deep learning has evolved so much that new techniques and algorithms spring every now and then. This paper examines several existing techniques that can be used for system log analysis.
Other Latest Articles
- Study and Analysis of Concrete Strength Parameters Using Red Mud as Partial Replacement of Binder Content with and without Hydrated Lime
- Social Entrepreneurship within Creative Industry: A Case Study in Music Committee
- Radiation Level in Western Part of Mongolia
- Twitter Bot Detection using Machine Learning Algorithms
- Thematic Approach for Enhancing Pedagogic Competency of Student Teachers at Secondary Level
Last modified: 2021-06-28 18:20:06