A5029926020- Advanced Graph Neural Networks
- Machine Learning in Materials Science
- Bioinformatics and Genomic Networks
- Computational Drug Discovery Methods
- Handwritten Text Recognition Techniques
- Data Mining Algorithms and Applications
- Anomaly Detection Techniques and Applications
- Time Series Analysis and Forecasting
- Advanced Computing and Algorithms
- Asymmetric Hydrogenation and Catalysis
- Intelligent Tutoring Systems and Adaptive Learning
- Semantic Web and Ontologies
- Innovative Microfluidic and Catalytic Techniques Innovation
- Graph Theory and Algorithms
- Bayesian Modeling and Causal Inference
- Chemical Synthesis and Analysis
- Educational Technology and Assessment
- Hand Gesture Recognition Systems
- Music and Audio Processing
Xi’an Jiaotong-Liverpool University
2022-2025
Correcting students' multiple-choice answers is a repetitive and mechanical task that can be considered an image multi-classification task. Assuming possible options are 'abcd' the correct option one of four, some students may write incorrect symbols or do not exist. In this paper, five classifications were set up - four for other writing. This approach takes into account possibility non-standard writing options.
In recent years, with the increasing use of educational technology and online learning platforms, there has been a growing interest in developing intelligent systems that can automatically predict knowledge points associated questions. This paper presents novel approach for point prediction middle school mathematics The dataset used this study consists large collection 591,379 To leverage power natural language processing techniques, questions are preprocessed using tokenizer encoded into...
Graph similarity measurement is a fundamental task in various graph-related applications. However, recent learning-based approaches lack interpretability as they directly transform interaction information between two graphs into hidden vector, making it difficult to understand how the score derived. To address this issue, we propose an end-to-end paradigm for graph learning called Similarity Computation via Maximum Common Subgraph Inference (INFMCS), which more interpretable. Our key insight...
Abstract Time series analysis is an important and challenging problem in data mining, where time a class of temporal objects. In the classification task, label dependent on features from last moments. Due to dependency, recurrent neural networks, as one prevalent learning-based architectures, take advantage relation among history data. The Long Short-Term Memory Network (LSTM) Gated Recurrent Unit (GRU) are two popular artificial networks used field deep learning. LSTM designed gate-like...
Graph similarity measurement, which computes the distance/similarity between two graphs, arises in various graph-related tasks. Recent learning-based methods lack interpretability, as they directly transform interaction information graphs into one hidden vector and then map it to similarity. To cope with this problem, study proposes a more interpretable end-to-end paradigm for graph learning, named Similarity Computation via Maximum Common Subgraph Inference (INFMCS). Our critical insight...
Predicting reactants from a specified core product stands as fundamental challenge within organic synthesis, termed retrosynthesis prediction. Recently, semi-template-based methods and graph-edits-based have achieved good performance in terms of both interpretability accuracy. However, due to their mechanisms these cannot predict complex reactions, e.g., reactions with multiple reaction center or attaching the same leaving group more than one atom. In this study we propose method,...
The reaction center consists of atoms in the product whose local properties are not identical to corresponding reactants. Prior studies on identification mainly semi-templated retrosynthesis methods. Moreover, they limited single identification. However, many centers comprised multiple bonds or reality. We refer it as center. This paper presents RCsearcher, a unified framework for and that combines advantages graph neural network deep reinforcement learning. critical insight this is must be...
Correcting students' multiple-choice answers is a repetitive and mechanical task that can be considered an image multi-classification task. Assuming possible options are 'abcd' the correct option one of four, some students may write incorrect symbols or do not exist. In this paper, five classifications were set up - four for other writing. This approach takes into account possibility non-standard writing options.
Retrosynthesis aims to break down desired molecules into accessible building blocks in a systematic manner. However, current template-based retrosynthesis approaches face limitations due fixed set of training templates, hindering their ability discover new chemical reactions. To overcome this challenge, we present novel prediction framework that can generate templates beyond the set. This innovative approach has demonstrated superior performance compared previous methods. Additionally,...