A5085691361- Topic Modeling
- Semantic Web and Ontologies
- Advanced Graph Neural Networks
- Natural Language Processing Techniques
- Biomedical Text Mining and Ontologies
- Data Quality and Management
- Domain Adaptation and Few-Shot Learning
- Multimodal Machine Learning Applications
- COVID-19 diagnosis using AI
- Text and Document Classification Technologies
- Machine Learning in Healthcare
- Time Series Analysis and Forecasting
- Advanced Text Analysis Techniques
- Bioinformatics and Genomic Networks
- Web Data Mining and Analysis
- Advanced Image and Video Retrieval Techniques
- Sentiment Analysis and Opinion Mining
- Obesity, Physical Activity, Diet
- Human Mobility and Location-Based Analysis
- Eating Disorders and Behaviors
- Computational Drug Discovery Methods
- Adversarial Robustness in Machine Learning
- Explainable Artificial Intelligence (XAI)
- Graph Theory and Algorithms
- Data Stream Mining Techniques
University of Manchester
2022-2025
University of Oxford
2018-2025
Chongqing Medical University
2024
eBay (United States)
2024
Bengbu Medical College
2022-2023
Nanchang University
2019-2020
Alibaba Group (United States)
2020
Science Oxford
2018-2020
Zhejiang University
2013-2016
Zhejiang Science and Technology Information Institute
2016
Deep neural networks have achieved promising results in stock trend prediction. However, most of these models two common drawbacks, including (i) current methods are not sensitive enough to abrupt changes trend, and (ii) forecasting interpretable for humans. To address problems, we propose a novel Knowledge-Driven Temporal Convolutional Network (KDTCN) prediction explanation. Firstly, extract structured events from financial news, utilize external knowledge graph obtain event embeddings....
Reasoning is essential for the development of large knowledge graphs, especially completion, which aims to infer new triples based on existing ones. Both rules and embeddings can be used graph reasoning they have their own advantages difficulties. Rule-based accurate explainable but rule learning with searching over always suffers from efficiency due huge search space. Embedding-based more scalable efficient as conducted via computation between embeddings, it has difficulty good...
Abstract Semantic embedding of knowledge graphs has been widely studied and used for prediction statistical analysis tasks across various domains such as Natural Language Processing the Web. However, less attention paid to developing robust methods OWL (Web Ontology Language) ontologies, which contain richer semantic information than plain graphs, have adopted in bioinformatics. In this paper, we propose a random walk word based ontology method named , encodes semantics an by taking into...
Automatically annotating column types with knowledge base (KB) concepts is a critical task to gain basic understanding of web tables. Current methods rely on either table metadata like name or entity correspondences cells in the KB, and may fail deal growing tables incomplete meta information. In this paper we propose neural network based type annotation framework named ColNet which able integrate KB reasoning lookup machine learning can automatically train Convolutional Neural Networks for...
Knowledge Graph Completion (KGC) has been proposed to improve Graphs by filling in missing connections via link prediction or relation extraction. One of the main difficulties for KGC is a low resource problem. Previous approaches assume sufficient training triples learn versatile vectors entities and relations, satisfactory number labeled sentences train competent extraction model. However, relations are very common KGs, those newly added often do not have many known samples training. In...
Zhiquan Ye, Yuxia Geng, Jiaoyan Chen, Jingmin Xiaoxiao Xu, SuHang Zheng, Feng Wang, Jun Zhang, Huajun Chen. Proceedings of the 58th Annual Meeting Association for Computational Linguistics. 2020.
Ontology alignment (a.k.a ontology matching (OM)) plays a critical role in knowledge integration. Owing to the success of machine learning many domains, it has been applied OM. However, existing methods, which often adopt ad-hoc feature engineering or non-contextual word embeddings, have not yet outperformed rule-based systems especially an unsupervised setting. In this paper, we propose novel OM system named BERTMap can support both and semi-supervised settings. It first predicts mappings...
Zero-shot learning (ZSL) aims to predict unseen classes whose samples have never appeared during training. One of the most effective and widely used semantic information for zero-shot image classification are attributes which annotations class-level visual characteristics. However, current methods often fail discriminate those subtle distinctions between images due not only shortage fine-grained annotations, but also attribute imbalance co-occurrence. In this paper, we present a...
In knowledge graph completion (KGC), predicting triples involving emerging entities and/or relations, which are unseen when the KG embeddings learned, has become a critical challenge. Subgraph reasoning with message passing is promising and popular solution. Some recent methods have achieved good performance, but they (i) usually can only predict alone, failing to address more realistic fully inductive situations both (ii) often conduct over relation patterns not utilized. this study, we...
Machine learning (ML), especially deep neural networks, has achieved great success, but many of them often rely on a number labeled samples for supervision. As sufficient training data are not always ready due to, e.g., continuously emerging prediction targets and costly sample annotation in real-world applications, ML with shortage is now being widely investigated. Among all these studies, prefer to utilize auxiliary information including those the form knowledge graph (KG) reduce reliance...
Large Language Models (LLMs) have taken Knowledge Representation -- and the world by storm. This inflection point marks a shift from explicit knowledge representation to renewed focus on hybrid of both parametric knowledge. In this position paper, we will discuss some common debate points within community LLMs (parametric knowledge) Graphs (explicit speculate opportunities visions that brings, as well related research topics challenges.
Knowledge Graphs (KGs) play a pivotal role in advancing various AI applications, with the semantic web community's exploration into multi-modal dimensions unlocking new avenues for innovation. In this survey, we carefully review over 300 articles, focusing on KG-aware research two principal aspects: KG-driven Multi-Modal (KG4MM) learning, where KGs support tasks, and Graph (MM4KG), which extends KG studies MMKG realm. We begin by defining MMKGs, then explore their construction progress. Our...
Integrating deep learning techniques, particularly language models (LMs), with knowledge representation techniques like ontologies has raised widespread attention, urging the need of a platform that supports both paradigms. Although packages such as OWL API and Jena offer robust support for basic ontology processing features, they lack capability to transform various types information within into formats suitable downstream learning-based applications. Moreover, widely-used APIs are...
As an emerging network paradigm, the Internet of Things (IoT) which consists a significant number multifunctional and heterogeneous IoT nodes has attracted dramatic attentions from both academia industry. With merits intelligent capacity, desirable scalability, high reliability, recently been applied for smart ocean applications to provide protection environment monitoring surveillance. Aiming coverage service border environmental surveillance, this article studies barrier problem...
Zero-shot Learning (ZSL), which aims to predict for those classes that have never appeared in the training data, has arisen hot research interests. The key of implementing ZSL is leverage prior knowledge builds semantic relationship between and enables transfer learned models (e.g., features) from (i.e., seen classes) unseen classes. However, priors adopted by existing methods are relatively limited with incomplete semantics. In this paper, we explore richer more competitive model...
Zero-shot learning (ZSL) which aims at predicting classes that have never appeared during the training using external knowledge (a.k.a. side information) has been widely investigated. In this paper we present a literature review towards ZSL in perspective of knowledge, where categorize their methods and compare different knowledge. With review, further discuss outlook role symbolic addressing other machine sample shortage issues.
Multi-modal entity alignment (MMEA) aims to discover identical entities across different knowledge graphs (KGs) whose are associated with relevant images. However, current MMEA algorithms rely on KG-level modality fusion strategies for multi-modal representation, which ignores the variations of preferences entities, thus compromising robustness against noise in modalities such as blurry images and relations. This paper introduces MEAformer, a mlti-modal transformer approach meta hybrid,...