News Archive

All academic staff and postgraduate students are encouraged to contribute to this proposed book " Ontology Learning and Knowledge Discovery Using the Web: Challenges and Recent Advances " edited by Mr Wilson Wong , Assistant Professor Wei Liu and Winthrop Professor Mohammed Bennamoun .

The book will be an invaluable resource as a library or personal reference for a wide ranging audience, including graduate students, researchers and industrial practitioners in the exciting field of ontology.

Due to the detailed scope and wide coverage of the book, it has the potential of being an upper-level course supplement for senior undergraduate students in Artificial Intelligence, and a resource for lecturers in Knowledge Acquisition, Knowledge Representation and Reasoning, Text Mining, Information Extraction, and Ontology Learning.

The deadline for proposals submission is 15 December 2009.

The proposed book details:

Introduction

Ontologies provide formal specifications of what might exist in a domain to ensure reusability and interoperability of multiple heterogeneous systems. Ontologies form an indispensable part of the Semantic Web standard stack. While the Semantic Web is still our vision into the future, ontologies have already found a myriad of applications such as document retrieval, question answering, image retrieval, agent interoperability and document annotation.

In recent years, automatic ontology learning from text has provided support and relief for knowledge engineers from the labourious task of manually engineering of ontologies. Ontology learning research, an area integrating advances from information retrieval, text mining, data mining, machine learning and natural language processing, has attracted increasing interests from a wide spectrum of application domains (e.g. bioinformatics, manufacturing). Being a rapidly growing area, it is crucial to collect the recent advances in tools and technologies in ontology learning and related areas.

Objective of the book

The main objective of this book is to provide relevant theoretical foundations, and disseminate new research findings and expert views on the remaining challenges in ontology learning.

In particular, the book focuses on the following questions:

• Can ontology learning continue to rely on techniques borrowed from related areas that were conceived for other purposes? Has the time arrived for us to look at certain peculiar requirements of ontology learning and develop specific techniques to meet these requirements?
• Lightweight ontologies are the most common type of ontologies in a variety of existing Semantic Web applications (e.g. knowledge management, document retrieval, communities of practice, data integration). Can these lightweight ontologies be easily extended to formal ones? If so, how?
• The poor coverage, rarity and maintenance cost related to manually-created resources such as semantic lexicons (e.g. WordNet, UMLS) and text corpora (e.g. BNC, GENIA corpus) have prompted an increasing number of researchers to turn to dynamic Web data for ontology learning. There is currently a lack of study concentrating on the systematic use of Web data as background knowledge for all phases of ontology learning. How do we know if we have the necessary background knowledge to carry out all our ontology learning tasks? Where do we look for more background knowledge if we know that what we have is inadequate?
• More and more practitioners in the domain of biology, health care, chemistry, manufacturing and others are looking up to ontology learning techniques for solutions to their knowledge sharing and reusability needs. How much more difficult is it to automatically learn ontologies from news articles, as compared to clinical notes or biomedical literature? To what extent can the current techniques meet the requirements of learning from texts across different domains? Is the field of automatic ontology learning from text ready for the industry?

Target Audience

This proposed book will be an invaluable resource as a library or personal reference for a wide range of audience, including, graduate students, researchers and industrial practitioners. Postgraduate students who are in the process of looking for future research directions, and carving out their own niche area will find this book particularly useful.

Due to the detailed scope and wide coverage of the book, it also has the potential of being an upper-level course supplement for senior undergraduate students in Artificial Intelligence, and a resource for lecturers in Knowledge Acquisition, Knowledge Representation and Reasoning, Text Mining, Information Extraction, and Ontology Learning.

Recommended topics include, but are not limited to:

Area 1: Text Processing

• Web data pre-processing
• Noisy text analytics
• Text annotation/Sentence parsing
• Textual content extraction/Boilerplates removal
• Automatic corpus construction

Area 2: Taxonomy Construction/Concept Formation

• Named entity recognition/noun phrase chunking
• Feature-based/featureless similarity and distance measures
• Term recognition/term extraction/terminology mining
• Cluster analysis/term clustering
• Entity disambiguation
• Relevance/contrastive analysis
• Latent semantic analysis
• Other machine learning-based techniques
• Other corpus-based techniques

Area 3: Relation and Axiom Discovery/Ontology Languages

• Lexico-syntactic patterns
• Use of dynamic Web data (e.g. Wikipedia mining, online dictionaries)
• Sub-categorisation frames
• Association rules mining
• Inductive logic programming
• Other corpus-based techniques
• Logic-based/frame-based/markup ontology languages

Further information is available on the website .

Media references

Mr Wilson Wong / [email protected] / 6488 2839
Assistant Professor Wei Liu / [email protected] / 6488 3095
Winthrop Professor Mohammed Bennamoun / [email protected] / 6488 2281