字符串数据异常检测算法比较

Outlier detection is a significant and well-researched problem in machine learning. However, there is limited research on outlier detection for string data, as most literature focuses on numerical data. A robust outlier detection algorithm for string data could aid in data cleaning or anomaly detection in system log files. This study compares two string outlier detection algorithms. Firstly, a variant of the well-known local outlier factor algorithm is introduced, tailored to detect outliers in string data using the Levenshtein measure to calculate the density of the dataset. A differently weighted Levenshtein measure is also presented, which takes into account hierarchical character classes and can be utilized to tune the algorithm for improved detection performance.