A5046118256- Topic Modeling
- Web Data Mining and Analysis
- Natural Language Processing Techniques
- Recommender Systems and Techniques
- Text Readability and Simplification
- Advanced Graph Neural Networks
- Sentiment Analysis and Opinion Mining
- Speech and dialogue systems
- Data Visualization and Analytics
- Information Retrieval and Search Behavior
- Machine Learning in Healthcare
- Online Learning and Analytics
- Intelligent Tutoring Systems and Adaptive Learning
- Advanced Text Analysis Techniques
Shanghai Jiao Tong University
2023-2024
Online education platforms, leveraging the internet to distribute resources, seek provide convenient but often fall short in real-time communication with students. They struggle offer personalized resources due challenge of addressing diverse obstacles students encounter throughout their learning journey. Recently, emergence large language models (LLMs), such as ChatGPT, offers possibility for resolving this issue by comprehending individual requests. Although LLMs have been successful...
Most click models focus on user behaviors towards a single list. However, with the development of interface (UI) design, layout displayed items result page tends to be multi-block style instead list, which requires different assumptions model more accurately. There exist for pages in desktop contexts, but they cannot directly applied mobile scenarios due interaction manners, types and especially presentation styles. In particular, can normally decomposed into interleavings basic vertical...
With the emergence of Large Language Models (LLMs), there has been a significant improvement in programming capabilities models, attracting growing attention from researchers. Evaluating LLMs is crucial as it reflects multifaceted abilities LLMs, and numerous downstream applications. In this paper, we propose CodeApex, bilingual benchmark dataset focusing on comprehension, code generation, correction LLMs. Programming comprehension task tests multiple-choice exam questions covering...
Utilizing large language models to generate codes has shown promising meaning in software development revolution. Despite the intelligence by general models, their specificity code generation can still be improved due syntactic gap and mismatched vocabulary existing among natural different programming languages. In addition, languages are inherently logical complex, making them hard correctly generated. Existing methods rely on multiple prompts model explore better solutions, which is...
Knowledge Tracing (KT) aims to determine whether students will respond correctly the next question, which is a crucial task in intelligent tutoring systems (ITS). In educational KT scenarios, transductive ID-based methods often face severe data sparsity and cold start problems, where interactions between individual questions are sparse, new concepts consistently arrive database. addition, existing models only implicitly consider correlation questions, lacking direct modeling of more complex...
Aspect-Opinion Pair Extraction (AOPE) and Aspect Sentiment Triplet (ASTE) have gained significant attention in natural language processing. However, most existing methods are a pipelined framework, which extracts aspects/opinions identifies their relations separately, leading to drawback of error propagation high time complexity. Towards this problem, we propose transition-based pipeline mitigate token-level bias capture position-aware aspect-opinion relations. With the use fused dataset...
To provide click simulation or relevance estimation based on users' implicit interaction feedback, models have been much studied during recent years. Most focus user behaviors towards a single list. However, with the development of interface (UI) design, layout displayed items result page tends to be multi-block (i.e., multi-list) style instead list, which requires different assumptions model more accurately. There exist for pages in desktop contexts, but they cannot directly applied mobile...