Author name cluster

Zongmin Ma

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

8 papers

1 author row

EAAI Journal 2026 Journal Article

Multi-hop reasoning of knowledge chain over fuzzy spatiotemporal knowledge graphs with box embeddings

Xiaowen Zhang
Li Yan
Zongmin Ma

Embedding models can map knowledge graphs into a vector space, transforming them into vector representations that machines can understand. This capability is essential for conducting knowledge reasoning over large-scale knowledge graphs. There have been some attempts to embed spatiotemporal knowledge graphs. While some efforts have explored spatiotemporal knowledge graph embeddings, real-world knowledge often involves uncertainty. This uncertainty arises not from nondeterminism in knowledge itself, but from incomplete information, varying source reliability, and imprecise linguistic expressions (e. g. , possibly, likely). To address this, uncertain spatiotemporal knowledge graphs (USTKGs)—which integrate spatial, temporal, and uncertainty dimensions—have gained increasing research attention. Consequently, embedding uncertain spatiotemporal knowledge graphs poses an inevitable challenge. Unfortunately, there is currently limited research on embedding uncertain spatiotemporal knowledge graphs. This paper discusses the modeling of uncertain spatiotemporal knowledge graphs in vector space and proposes an Uncertain Spatiotemporal Knowledge Graph Embedding model (USTKGE). USTKGE can map uncertain spatiotemporal knowledge graphs entirely into vector space, achieving independent representations of elements of uncertain spatiotemporal knowledge tuples in vector space. We designed two logical constraints to enhance the robustness of USTKGE. Furthermore, to perform multi-hop reasoning over uncertain spatiotemporal knowledge graphs, we investigate a knowledge chain multi-hop reasoning method based on USTKGE. Experimental results demonstrate the rationality and effectiveness of this method.

Details DOI

EAAI Journal 2026 Journal Article

Root cause analysis of cloud platform faults based on anomaly correlation graph and graph neural network

Beijing Zhou
Bingyan Chen
Li Yan
Xinyu Liang
Zongmin Ma

The increasingly complex structure of the cloud environment makes it more difficult to deal with faults. Fault root cause analysis can locate the root cause of fault and promote the troubleshooting process. Traditional fault root cause analysis usually relies on the experienced judgment of engineers, which is time-consuming and poor in scalability. This paper proposes a novel model to solve the problem of fault root cause analysis in cloud platforms with complex network topology and massive alarms generation. Our model combines anomaly correlation graph, graph attention network and light gradient boosting machine. The causal correlations are first learned from historical alarm data to construct the anomaly correlation graph. To reduce noise interference, we propose a weight threshold sampling mechanism to select neighbor nodes with high causal correlation as the aggregation targets of the central node. Graph attention network is then used to aggregate neighborhood information to achieve node feature extraction, which can effectively represent the complex dependencies among nodes. Finally, light gradient boosting machine is applied for classification to realize root cause localization. Moreover, interpretable reports are generated through similarity analysis to provide references for engineers in repairing faults. The model can locate the root cause without excessive mass training samples. The experimental results show that our proposed model can achieve better effect of root cause analysis than other relevant methods, where the precision is increased by more than 10%.

Details DOI

EAAI Journal 2026 Journal Article

Spatiotemporal knowledge graph multi-hop reasoning based on large language models

Xinyu Liang
Xiaobei Xu
Ruizhe Ma
Li Yan
Zongmin Ma

With the continuous progress of artificial intelligence, the application of large language models (LLMs) has provided completely new possibilities for knowledge graph reasoning and semantic understanding. The proposed method focuses on solving the multi-hop link prediction problem in spatiotemporal knowledge graphs through reasoning. For this purpose, an innovative framework Spatiotemporal Graph Multi-hop Reasoning Based on Large Language Model (STMH-LLM) is proposed. This method transforms the structured knowledge graph data into natural language descriptions and utilizes these prompts to fine-tuning LLMs, thereby enhancing the performance of multi-step relational inference. The proposed framework aims to capture the latent representations of entities and their relationship networks through natural language prompts. To verify the effectiveness of STMH-LLM, the commonly used open-source large language models Qwen2. 5 and GLM4 were selected for fine-tuning tests. In addition, this framework also demonstrates the potential to provide zero-shot reasoning capabilities for LLMs, enabling them to handle previously unseen prompts. Experiments were conducted on three datasets, and the results show that, compared with traditional models, STMH-LLM significantly improves the generalization ability of the model and achieves more accurate predictions in unfamiliar scenarios.

Details DOI

EAAI Journal 2025 Journal Article

Correcting update anomalies in spatiotemporal knowledge graphs with uncertainties

Xiaowen Zhang
Li Yan
Beijing Zhou
Zongmin Ma

As a knowledge description framework certified by the W3C (World Wide Web Consortium), Resource Description Framework (RDF) has a strict yet flexible description syntax and is widely accepted as a carrier of knowledge graphs (KGs). Currently, a large amount of research work has been done on static knowledge graphs. However, dealing with dynamic knowledge graphs, especially spatiotemporal knowledge, remains an important research topic. Knowledge in the real world is not always deterministic. Consequently, some research on uncertain knowledge modeling and management has been proposed. This paper focuses on fuzzy spatiotemporal knowledge modeling within the context of the RDF model. We formally propose an extended RDF model to represent fuzzy spatiotemporal knowledge. This model can represent the continuous motion trajectory of fuzzy spatiotemporal entities and avoid repeated time intervals and location information in RDF. Specifically, we systematically identify the consistency constraints in the fuzzy spatiotemporal RDF model. Based on these constraints, we further propose a method for correcting timely inconsistencies when updating fuzzy RDF graph data. Experimental results demonstrate the effectiveness of the fuzzy spatiotemporal RDF model and the inconsistency correction method proposed in this paper.

Details DOI

EAAI Journal 2025 Journal Article

DiffAT: Effective data augmentation with diffusion models for time series forecasting

Yang Yu
Ruizhe Ma
Wenbo Gu
Zongmin Ma

Data augmentation offers a promising solution to data scarcity in deep learning-based time series forecasting. However, current approaches face dual limitations (1) Hand-designed methods (e. g. , cropping/masking): often disrupt the continuity of vital temporal patterns (such as seasonal and trend) by introducing abrupt pattern discontinuities; (2) Generative models often face difficulties in preserving task-critical features that are essential for prediction, especially when aiming to generate diverse augmented series. To tackle these dilemmas, we propose a novel conditional diffusion-based data augmentation framework, named DiffAT, for time series forecasting tasks. DiffAT synergizes: (1) Patch-wise masking reconstruction to capture structural invariants (such as autocorrelation and causality), and (2) encoding hand-designed augmentation prototypes for guiding diversity-preserving generation. DiffAT achieves dual enhancement: maintaining continuity of temporal patterns through progressive denoising process and exposing latent invariant patterns via guided diversity injection. We validate the efficacy of DiffAT through extensive experiments on seven real-world datasets, by comparing DiffAT with six state-of-the-art time series data augmentation methods. The results indicate our method can boost the forecasting performance of Autoformer by up to 6. 49 % in 26/28 cases and improve forecasting performance of LightTS by up to 3. 11 % in 23/28 cases on 7 real-world benchmarks. Extensive experiments also indicate that DiffAT can improve the accuracy of forecasting models in few-shot scenario (with 1 % training data) in 54/60 cases. We will release the source code upon publication.

Details DOI

KER Journal 2016 Journal Article

A literature overview of knowledge sharing between Petri nets and ontologies

Haitao Cheng
Zongmin Ma

Abstract Knowledge representation is a subarea of artificial intelligence concerned with using formal symbols to represent a set of facts within a knowledge domain. Two popular knowledge representation languages, namely Petri net and ontology, are promising knowledge sharing and reusing methods in knowledge engineering. The combination of Petri net and ontology can facilitate achieving complementary advantages. Currently, many efforts have been done on knowledge sharing between Petri nets and ontologies. To investigate these issues and more importantly serve as identifying the direction of knowledge sharing between Petri nets and ontologies, in this paper we give a comprehensive literature overview of knowledge sharing between Petri net models and ontology models to satisfy the obvious need. In detail, we discuss the knowledge sharing from two aspects: the different knowledge representation approaches of ontology to represent and reason Petri net and issues of constructing Petri net from ontology. In addition, other important issues on applications and directions for future research are discussed in detail.

Details DOI

KER Journal 2016 Journal Article

A survey on fuzzy ontologies for the Semantic Web

Fu Zhang
Jingwei Cheng
Zongmin Ma

Abstract Ontology, as a standard (World Wide Web Consortium recommendation) for representing knowledge in the Semantic Web, has become a fundamental and critical component for developing applications in different real-world scenarios. However, it is widely pointed out that classical ontology model is not sufficient to deal with imprecise and vague knowledge strongly characterizing some real-world applications. Thus, a requirement of extending ontologies naturally arises in many practical applications of knowledge-based systems, in particular the Semantic Web. In order to provide the necessary means to handle such vague and imprecise information there are today many proposals for fuzzy extensions to ontologies, and until now the literature on fuzzy ontologies has been flourishing. To investigate fuzzy ontologies and more importantly serve as helping readers grasp the main ideas and results of fuzzy ontologies, and to highlight an ongoing research on fuzzy approaches for knowledge semantic representation based on ontologies, as well as their applications on various domains, in this paper, we provide a comprehensive overview of fuzzy ontologies. In detail, we first introduce fuzzy ontologies from the most common aspects such as representation (including categories, formal definitions, representation languages, and tools of fuzzy ontologies), reasoning (including reasoning techniques and reasoners), and applications (the most relevant applications about fuzzy ontologies). Then, the other important issues on fuzzy ontologies, such as construction, mapping, integration, query, storage, evaluation, extension, and directions for future research, are also discussed in detail. Also, we make some comparisons and analyses in our whole review.

Details DOI

KER Journal 2016 Journal Article

Storing massive Resource Description Framework (RDF) data: a survey

Zongmin Ma
Miriam A. M. Capretz
Li Yan

Abstract The Resource Description Framework (RDF) is a flexible model for representing information about resources on the Web. As a W3C (World Wide Web Consortium) Recommendation, RDF has rapidly gained popularity. With the widespread acceptance of RDF on the Web and in the enterprise, a huge amount of RDF data is being proliferated and becoming available. Efficient and scalable management of RDF data is therefore of increasing importance. RDF data management has attracted attention in the database and Semantic Web communities. Much work has been devoted to proposing different solutions to store RDF data efficiently. This paper focusses on using relational databases and NoSQL (for ‘not only SQL (Structured Query Language)’) databases to store massive RDF data. A full up-to-date overview of the current state of the art in RDF data storage is provided in the paper.

Details DOI