| |
Log-driven Intelligent Software Reliability Engineering
Speaker (s):
HUO Yintong
PhD Candidate
Computer Science & Engineering
The Chinese University of Hong Kong
|
|
Date:
Time:
Venue:
|
|
08 January 2024, Monday
10:00am – 11:15am
School of Computing & Information
Systems 1 (SCIS 1)
Level 5, Meeting Room 5-1
Singapore Management University
80 Stamford Road, Singapore 178902
Please register by 04 January 2024
We look forward to seeing you at this seminar.
|
|
About the Talk
Modern software systems are serving many aspects of our daily activities, from search engines to communication platforms. It is important to ensure their reliability. The big data era has introduced machine learning (ML) for Software Reliability Engineering (SRE), offering promising opportunities. In this talk, the speaker will present her insights and work on intelligent SRE, with a focus on utilizing log data for fault prevention, fault removal, and fault tolerance phase.
To begin with, she will discuss SemParser, a semantic-aware log parser for log analysis. As the fundamental step of log analysis, log parsing aims to extract structured templates and parameters from raw log messages. Instead of previous syntax-based parsers, SemParser extracts semantics through a two-step process: 1) pairing parameters and their descriptions within individual logs, and 2) inferring cross-log semantics and computing structural outputs.
Next, she will present AutoLog, a source code-guided log sequence generator. Log analysis grapples with a significant challenge—insufficient practical data due to privacy constraints in industrial logs and the simplistic nature of publicly available lab logs. AutoLog tackles this issue by leveraging program analysis techniques to simulate log sequences based on the execution order of logging statements.
Her past research showcased the effectiveness of log-driven methods in advancing SRE. Lastly, she will demonstrate her research plan in the next 2-5 years, stemming from intelligent log operations and expanding into diverse applications in software development.
About the Speaker
HUO Yintong is currently a Ph.D. candidate at the Chinese University of Hong Kong, advised by Michael R. Lyu. Yintong obtained her Bachelor’s Degree at the University of Electronic Science and Technology of China. Her research area is intelligent Software Engineering (SE), which applies Natural Language Processing (NLP) techniques to facilitate software development, testing, and maintenance processes. She has published 9 papers in top-tier SE and NLP conferences. She is the recipient of an IEEE Open Software Services Award for the LogPAI project (3k+ Stars, 70k+ Downloads). Explore further details on her webpage: https://yintonghuo.github.io/yintonghuo/.
She is a tenure-track faculty candidate for the Information Systems & Technology, Software Engineering cluster.
|
