Wang Wei

wei's picture

Centre for Communication Systems Research,
Faculty of Engineering and Physical Sciences,
University of Surrey,
Guildford, Surrey, GU2 7XH,
United Kingdom

Phone: +44 1483 68 3423
Fax: +44 1483 68 6011


I am a research fellow at the Centre for Communication and Systems Research at the University of Surrey. I obtained my PhD in Computer Science from University of Nottingham in 2009 and BSc in Computer Science from the same university in 2006. Before joining the University of Surrey, I worked as an Assistant Professor at the University of Nottingham Malaysia Campus. My research interests include semantic search, ontology learning from unstructured text, knowledge acquisition, social recommender systems, semantic analysis on the social web, semantic web service discovery and matchmaking, data mining, and machine learning.


Internet of Things Environment for Service Creation and Testing
IoT.est will develop a test-driven service creation environment (SCE) for Internet of Things enabled business services. The SCE will enable the acquisition of data and control/actuation of sensors, objects and actuators. The project will provide the means and tools to define and instantiate IoT services that exploit data across domain boundaries and facilitate run-time monitoring which enables autonomous service adaptation to environment/context and network parameter (e.g. QoS) changes. The project will prototype its major concepts and will evaluate the results for exploitation towards future IoT service creation, deployment and testing products.

IRIS2: A Semantic Search Engine that Does Rational Research
Dominant ranking paradigms used in today's search engines for scientific publications include those based on content or link analysis, or combination of the both, which have their foundations in development of the modern information retrieval. Information objects and users in the scientific world have their own characteristics, however, they have not been sufficiently exploited in those existing ranking methods. We present the IRIS2 semantic search engine, which utilises a ranking method based on the ``rational research'' model, for searching and ranking entities related to scientific publications. In IRIS2, entities in a schema ontology and a knowledge base model information objects and semantic relationships among those entities simulate researcher's activities. The resulting ranking algorithm restores an elegant idea that a researcher does rational research in a scientific research environment.

Rational Research Model for Ranking Semantic Entities
Ranking plays important roles in contemporary Internet and vertical search engines. Among existing ranking algorithms, link analysis based algorithms have been proved as effective means for ranking retrieved documents from large-scale text repositories, such as the current Web. Recent development in semantic Web and semantic search raises considerable interests in new ranking paradigms for various applications. While ranking methods in this context exist, they have not gained much popularity. In this article we propose the RareRank algorithm for ranking entities in semantic search applications. The algorithm is based on the ``Rational Research'' model which reflects search behaviour of a ``rational'' researcher in a scientific research environment. Justification, design, and implementation of the algorithm are elaborated in details. In the experiment, the algorithm is deployed for ranking semantic entities, and results were evaluated by domain experts using popular ranking performance measures. A comparison study with existing link-based ranking algorithms reveals the preponderance of the proposed method.

Automated Terminological Ontology Learning
Probabilistic topic models were originally developed and utilised for document modeling and topic extraction in Information Retrieval. In this paper we describe a new approach for automatic learning of terminological ontologies from text corpus based on such models. In our approach, topic models are used as efficient dimension reduction techniques, which are able to capture semantic relationships between word-topic and topic-document interpreted in terms of probability distributions. We propose two algorithms for learning terminological ontologies using the principle of topic relationship and exploiting information theory with the probabilistic topic models learned. Experiments with different model parameters were conducted and learned ontology statements were evaluated by the domain experts. We have also compared the results of our method with two existing concept hierarchy learning methods on the same dataset. The study shows that our method outperforms other methods in terms of recall and precision measures. The precision level of the learned ontology is sufficient for it to be deployed for the purpose of browsing, navigation, and information search and retrieval in digital libraries.


I am currently involved in the following projects:


Journal Articles
Book Chapter
Conference Papers


Journal Reviewer
Conference Technical Programme Committe Member
Various Events


album-1 | album-2