Log Message Anomaly Detection and Classification Using Auto-B/LSTM and Auto-GRU
This addresses the challenge of classifying millions of unstructured daily logs in software systems, though it appears incremental as it applies hybrid deep learning methods to a known problem.
The paper tackled log message anomaly detection and classification by developing Auto-LSTM, Auto-BLSTM, and Auto-GRU models to convert unstructured log data into features, achieving better performance than other well-known algorithms on datasets like BGL, Openstack, Thunderbird, and IMDB.
Log messages are now widely used in software systems. They are important for classification as millions of logs are generated each day. Most logs are unstructured which makes classification a challenge. In this paper, Deep Learning (DL) methods called Auto-LSTM, Auto-BLSTM and Auto-GRU are developed for anomaly detection and log classification. These models are used to convert unstructured log data to extracted features which is suitable for classification algorithms. They are evaluated using four data sets, namely BGL, Openstack, Thunderbird and IMDB. The first three are popular log data sets while the fourth is a movie review data set which is used for sentiment classification. The results obtained show that Auto-LSTM, Auto-BLSTM and Auto-GRU perform better than other well-known algorithms.