Xiaolin Chai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiaolin Chai has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Yan Sun
School of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing, China
6 shared publications - Mohsen Guizani
Mohamed bin Zayed University of Artificial Intelligence
3 shared publications - Hong Luo
School of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing, China
2 shared publications - Aobo Liang
School of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing, China
1 shared publication - Ce Shang
Department of Computer Science, Beijing University of Posts and Telecommunications, Beijing, China
1 shared publication Hang Zhang
Department of Computer Science, Beijing University of Posts and Telecommunication, Beijing, China
1 shared publication
Latest Publications
- 14 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (4) , 1473-1487
RELog: Robust and Efficient Anomaly Detection Based on Complete Logs for Large-Scale Systems
As large-scale systems continue to grow in complexity, effectively leveraging the multi-dimensional information contained in semi-structured logs has become a critical challenge for ensuring system stability. Existing methods rely on limited log content, which restricts semantic modeling capability and leads to degraded performance when abnormal samples are scarce or data distributions are imbalanced. To address these challenges, we propose RELog, an efficient and robust anomaly detection framework that comprehensively utilizes multi-dimensional log features. RELog adopts a hierarchical process for efficient log parsing and anomaly detection. During log parsing, Large Language Models are introduced to enhance the semantic understanding of low-confidence templates generated by the heuristic method. For anomaly detection, we design dedicated encoders for templates, parameters, and time features, tailored to the low complexity and high redundancy of logs, enabling discriminative embeddings. Furthermore, we propose an efficient hybrid sequence encoder integrating state-space modeling and attention for capturing continuous log patterns and critical dependencies, complemented by a replacement classification task for imbalanced data. Experimental results demonstrate that RELog achieves high-accuracy log parsing and anomaly detection while maintaining efficiency. Moreover, RELog demonstrates robustness across varying anomaly ratios and dynamic environments, effectively detecting diverse types of anomalies reflected by parameter and time features.
- 14 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (4) , 1473-1487
RELog: Robust and Efficient Anomaly Detection Based on Complete Logs for Large-Scale Systems
As large-scale systems continue to grow in complexity, effectively leveraging the multi-dimensional information contained in semi-structured logs has become a critical challenge for ensuring system stability. Existing methods rely on limited log content, which restricts semantic modeling capability and leads to degraded performance when abnormal samples are scarce or data distributions are imbalanced. To address these challenges, we propose RELog, an efficient and robust anomaly detection framework that comprehensively utilizes multi-dimensional log features. RELog adopts a hierarchical process for efficient log parsing and anomaly detection. During log parsing, Large Language Models are introduced to enhance the semantic understanding of low-confidence templates generated by the heuristic method. For anomaly detection, we design dedicated encoders for templates, parameters, and time features, tailored to the low complexity and high redundancy of logs, enabling discriminative embeddings. Furthermore, we propose an efficient hybrid sequence encoder integrating state-space modeling and attention for capturing continuous log patterns and critical dependencies, complemented by a replacement classification task for imbalanced data. Experimental results demonstrate that RELog achieves high-accuracy log parsing and anomaly detection while maintaining efficiency. Moreover, RELog demonstrates robustness across varying anomaly ratios and dynamic environments, effectively detecting diverse types of anomalies reflected by parameter and time features.
- 1 March 2025
- Published by Springer Nature
- p. 168-177
Generalizable and Robust Log Anomaly Detection Based on Transformer
- 1 March 2025
- Published by Springer Nature
- p. 168-177
Generalizable and Robust Log Anomaly Detection Based on Transformer
- 16 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (2) , 1150-1167
Log Sequence Anomaly Detection Based on Template and Parameter Parsing via BERT
Logs record various operations and events during system running in text format, which is an essential basis for detecting and identifying potential security threats or system failures, and is widely used in system management to ensure security and reliability. Existing log sequence anomaly detection is limited by log parsing and does not consider all key features of logs, which may cause false or missed detection. In this paper, we propose a fast and accurate log parsing method and feed the entire log content into the deep learning network for analysis. To avoid semantic loss during parsing, we replace some variables with tokens containing semantic information and divide logs with appropriate granularity. To ensure the speed and accuracy of parsing, we propose a similarity-based fast merging method to deal with redundant templates. For anomaly detection, we use the complete log content features as input to the model. We use Bidirectional Encoder Representation from Transformers (BERT) to output anomaly detection results directly after considering both the global and local information of log sequences. Experiments show that our log parsing method achieves the best average parsing quality on 16 datasets, and the anomaly detection method achieves optimal results on different datasets.
- 16 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (2) , 1150-1167
Log Sequence Anomaly Detection Based on Template and Parameter Parsing via BERT
Logs record various operations and events during system running in text format, which is an essential basis for detecting and identifying potential security threats or system failures, and is widely used in system management to ensure security and reliability. Existing log sequence anomaly detection is limited by log parsing and does not consider all key features of logs, which may cause false or missed detection. In this paper, we propose a fast and accurate log parsing method and feed the entire log content into the deep learning network for analysis. To avoid semantic loss during parsing, we replace some variables with tokens containing semantic information and divide logs with appropriate granularity. To ensure the speed and accuracy of parsing, we propose a similarity-based fast merging method to deal with redundant templates. For anomaly detection, we use the complete log content features as input to the model. We use Bidirectional Encoder Representation from Transformers (BERT) to output anomaly detection results directly after considering both the global and local information of log sequences. Experiments show that our log parsing method achieves the best average parsing quality on 16 datasets, and the anomaly detection method achieves optimal results on different datasets.
- 27 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (19) , 31467-31478
GTformer: Graph-Based Temporal-Order-Aware Transformer for Long-Term Series Forecasting
In the production environment of the Internet of Things (IoT), sensors of various qualities generate a large amount of multivariate time series (MTS) data. The long-term prediction of time series data generated by various IoT devices provides longer foresight and help execute necessary resource scheduling or fault alarms in advance, thus improving the efficiency of system operation and ensuring system security. In recent years, deep learning models like Transformers have achieved advanced performance in multivariate long-term time series forecasting (MLTSF) tasks. However, many previous research attempts either overlook the inter-series dependencies or ignore to model strict temporal order of MTS data. In this paper, we introduce GT-former, a graph-based temporal-order-aware transformer model. We propose an adaptive graph learning method specifically designed for MTS data to capture both uni-directional and bi-directional relations. In addition, we generate positional encoding in a sequential way to emphasize the strict temporal order of time series. By adopting these two components, our model can have a better understanding of the inter-series and intra-series dependencies of MTS data. We conduct extensive experiments on eight real-world datasets and the results show that our model achieves better predictions compared with state-of-the-art methods.
- 27 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (19) , 31467-31478
GTformer: Graph-Based Temporal-Order-Aware Transformer for Long-Term Series Forecasting
In the production environment of the Internet of Things (IoT), sensors of various qualities generate a large amount of multivariate time series (MTS) data. The long-term prediction of time series data generated by various IoT devices provides longer foresight and help execute necessary resource scheduling or fault alarms in advance, thus improving the efficiency of system operation and ensuring system security. In recent years, deep learning models like Transformers have achieved advanced performance in multivariate long-term time series forecasting (MLTSF) tasks. However, many previous research attempts either overlook the inter-series dependencies or ignore to model strict temporal order of MTS data. In this paper, we introduce GT-former, a graph-based temporal-order-aware transformer model. We propose an adaptive graph learning method specifically designed for MTS data to capture both uni-directional and bi-directional relations. In addition, we generate positional encoding in a sequential way to emphasize the strict temporal order of time series. By adopting these two components, our model can have a better understanding of the inter-series and intra-series dependencies of MTS data. We conduct extensive experiments on eight real-world datasets and the results show that our model achieves better predictions compared with state-of-the-art methods.
- 23 August 2022
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 37 (11) , 9598-9619
An artistic analysis model based on sequence cartoon images for scratch
- 23 August 2022
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 37 (11) , 9598-9619
An artistic analysis model based on sequence cartoon images for scratch
- 13 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Emerging Topics in Computing
- Vol. PP (2) , 1
DWES: A Dynamic Weighted Evaluation System for Scratch based on Computational Thinking
Scratch is a programming environment, which is widely adopted as the first language to enter the programming world. Since Scratch projects are rich in multimedia resources, it can be time-consuming to judge them by hand, requiring manual execution with the understanding of each project. Auto-judgment tools come into being, they can rapidly evaluate the project in some way. Unfortunately, the existing tools are too rigid and they ignore the diversity of Scratch projects. To address this issue, we propose a new dynamic weighted evaluation system (DWES). First, we define a new computational thinking (CT) evaluation criteria from eight aspects. Second, based on the proposed criteria, we present an analysis tool that automatically assesses the CT skills of the learners' projects. Third, considering the characteristics of projects, we dynamically adjust the evaluation results according to types, so that a single standard is no longer applied roughly to all. This process is data-driven, we consider the CT performance and scripts of projects. We prove the rationality of the evaluation criteria from the aspect of program complexity. Through the correlation analysis between DWES CT scores and experts' ratings, we find that compared with Dr. Scratch, the correlation coefficient has increased.
- 13 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Emerging Topics in Computing
- Vol. PP (2) , 1
DWES: A Dynamic Weighted Evaluation System for Scratch based on Computational Thinking
Scratch is a programming environment, which is widely adopted as the first language to enter the programming world. Since Scratch projects are rich in multimedia resources, it can be time-consuming to judge them by hand, requiring manual execution with the understanding of each project. Auto-judgment tools come into being, they can rapidly evaluate the project in some way. Unfortunately, the existing tools are too rigid and they ignore the diversity of Scratch projects. To address this issue, we propose a new dynamic weighted evaluation system (DWES). First, we define a new computational thinking (CT) evaluation criteria from eight aspects. Second, based on the proposed criteria, we present an analysis tool that automatically assesses the CT skills of the learners' projects. Third, considering the characteristics of projects, we dynamically adjust the evaluation results according to types, so that a single standard is no longer applied roughly to all. This process is data-driven, we consider the CT performance and scripts of projects. We prove the rationality of the evaluation criteria from the aspect of program complexity. Through the correlation analysis between DWES CT scores and experts' ratings, we find that compared with Dr. Scratch, the correlation coefficient has increased.
- 1 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-2
A New Evaluation Metrics for Block-based Python Code
The programming education for the students set off a global enthusiasm. The Block-Python programming tools can fill the gap from block-based programming to language coding. However, there is little work focus on the evaluation metrics for block-based python code. The past block-oriented evaluation metrics based on Computational Thinking (CT) were not language-oriented, and program volume was often ignored. In this paper, we propose a new evaluation metrics for block-based python code, which combines the CT dimensions and program volume. The experimental results show that the correlations between the proposed metrics and Halstead, McCabe complexity algorithms are both above 0.7, which is higher than Dr. Scratch.
- 1 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-2
A New Evaluation Metrics for Block-based Python Code
The programming education for the students set off a global enthusiasm. The Block-Python programming tools can fill the gap from block-based programming to language coding. However, there is little work focus on the evaluation metrics for block-based python code. The past block-oriented evaluation metrics based on Computational Thinking (CT) were not language-oriented, and program volume was often ignored. In this paper, we propose a new evaluation metrics for block-based python code, which combines the CT dimensions and program volume. The experimental results show that the correlations between the proposed metrics and Halstead, McCabe complexity algorithms are both above 0.7, which is higher than Dr. Scratch.