Hao Liu

Hao Liu

Auto-generated profile, based on publicly available information

Hao Liu

School of Software Engineering, South China University of Technology, Guangzhou, 510006, Guangdong, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 December 2025 in Information Processing & Management
Verified
Journal Article
Closed access
HLHao Liu
Hao Liu
DLDong Li
Dong Li
BZBing Zeng
Bing Zeng
HRHaopeng Ren
Haopeng Ren

1 October 2023
Verified
Conference Paper
Closed access
XZXiang Zhang
Xiang Zhang
TSTao Sun
Tao Sun
ZZZihao Zhang
Zihao Zhang
XYXinyan Yin
Xinyan Yin
HZHongyan Zheng
Hongyan Zheng
HLHao Liu
Hao Liu
Socialmedia is highly active and full of rumors while the information is widely propagation. Mastering the process information of rumor propagation plays an important role in promoting automated rumor detection. However, the existing methods ignore the rumor evolution and development process, resulting in the loss of details, which is not con-ducive to extracting key information in rumor detection. In this paper, we propose a novel hierarchical dynamic graph convolution network (H-DynGCN-Enh) to build a dynamic propagation graph for each news differentiation, while taking into account the bi-directional features of rumor propagation and dissemination, fully extracting the key information. In addition, we introduce attention-based graph pooling to design a feature enhancement strategy to avoid redundant information and increase the interaction between comments and original news. Finally, we use the multi-head self-attention mechanism to adjust the weight of dynamic information to obtain global information for detection. The results of two open real-world datasets demonstrate that our model has exceeded the most advanced baseline, the effectiveness of each module has also been fully proved in the ablation experiment.

21 August 2025 in ACM Transactions on Knowledge Discovery From Data
Verified
Journal Article
Open access
HLHao Liu
Hao Liu
DLDong Li
Dong Li
BZBing Zeng
Bing Zeng
WLWei Liang
Wei Liang
DLDongjie Li
Dongjie Li
Knowledge Graphs (KGs) are extensively used in recommendation systems and information retrieval but often suffer from incompleteness. A popular solution to this problem is multi-hop inference through a reinforcement learning framework, which provides an interpretable path for predicting missing links in KGs. Most previous work focuses on improving the performance of multi-hop link prediction. However, it has been observed that many multi-hop paths generated by these methods are irrational; they often fail to reasonably explain the predicted answer entities. To address this challenge, we introduce the Joint Multi-hop Link Prediction (JMLP) framework. The framework consists of a relation attention network and an entity attention network, which collaboratively generate the reasoning paths. The relation attention module utilizes an induction network to encode historical paths and employs the graph self-attention mechanism to refine the interaction of relation contextual information. The entity attention module uses the graph attention mechanism to obtain the aggregated contextual features and leverages self-attention to strengthen the correlation between local and global contextual entity features. Extensive experiments on five datasets validate the effectiveness of our approach, demonstrating significant improvements both in predictive performance and interpretability compared to state-of-the-art methods.

25 July 2025
Verified
Book Chapter
Closed access
PDF file
HLHao Liu
Hao Liu
NLNingsi Li
Ningsi Li
SLSheng Liu
Sheng Liu
LLLulu Liu
Lulu Liu

15 November 2025 in Neural Networks
Verified
Journal Article
Closed access
HLHao Liu
Hao Liu
DLDong Li
Dong Li
BZBing Zeng
Bing Zeng
YXYang Xu
Yang Xu

20 October 2023
Verified
Conference Paper
Closed access
YLYang Liu
Yang Liu
HLHao Liu
Hao Liu
Wei ChenWei Chen
Wei Chen
The current intent recognition method has insufficient generalization ability, and the evaluation of confidence level of intent recognition is the key to its popularization and application in the high-reliability military and civilian fields. The spatiotemporal Knowledge Graph can uniformly represent and efficiently organize entities containing spatial location information and temporal information. Battlefield situational objectives and their relationships can be abstracted as a triple composed of two specific types of spatiotemporal entities and combat intent relationships. Using typical confrontation scene data to train neural networks, and fusing the confidence levels of the entity and Knowledge Graph, the confidence level evaluation results of the intent triple for the existing spatiotemporal knowledge base are calculated. The simulation test results show that the method can effectively evaluate the confidence level of the intent recognition results, and the accuracy rate is improved compared with the current mainstream methods. This method breaks the "closure" assumption of the Knowledge Graph update and completion algorithm from the algorithm idea level, and completes the last link in the popularization and application of the intent recognition method