We review the use on ontologies for the integration of heterogeneous information sources. Based on an in-depth evaluation of existing approaches to this problem we discuss how ontologies are used to support the integration task. We evaluate and compare the languages used to represent the ontologies and the use of mappings between ontologies as well as to connect ontologies with information sources. We also enquire into ontology engineering methods and tools used to develop ontologies for information integration. Based on the results of our analysis we summarize the state-of-the-art in ontology-based information integration and name areas of further research activities.

This is the second of a two-part paper to review ontology research and development, in particular, ontology mapping and evolving. Ontology is defined as a formal explicit specification of a shared conceptualization. Ontology itself is not a static model so that it must have the potential to capture changes of meanings and relations. As such, mapping and evolving ontologies is part of an essential task of ontology learning and development. Ontology mapping is concerned with reusing existing ontologies, expanding and combining them by some means and enabling a larger pool of information and knowledge in different domains to be integrated to support new communication and use. Ontology evolving, likewise, is concerned with maintaining existing ontologies and extending them as appropriate when new information or knowledge is acquired. It is apparent from the reviews that current research into semi-automatic or automatic ontology research in all the three aspects of generation, mapping and evolving have so far achieved limited success. Expert

Abstract. In this paper we present a methodology to extract information from the Web to build a taxonomy of terms and Web resources for a given domain. This taxonomy represents a hierarchy of classes and gives to the user a general view of the kind of concepts and the most significant sites that he can find on the Web for the specified domain. The system uses intensively a publicly available search engine, extracts concepts (based on its relation to the initial one and statistical data about appearance), selects and categorizes the most representative Web resources of each one and represents the result in a standard way. 1

In recent years there have been some efforts to automate the ontology acquisition and construction process. The proposed systems differ from each other in some distinguishing factors and have many features in common. This paper presents the state of the art in ontology learning (OL) and introduces a framework for classifying and comparing OL systems. The dimensions of the framework answer to questions about what to learn, from where to learn and how to learn. They include features of the input, the methods of learning and knowledge acquisition, the elements learned, the resulted ontology and also the evaluation process. To extract the framework over 50 OL systems or modules from the recent workshops, conferences and published journals are studied and seven prominent of them with most differences are selected to be compared according to our framework. In this paper after a brief description of the seven selected systems we will describe the framework dimensions. Then we will place the representative ontology learning systems into our framework. At last we will describe the differences, strengths and weaknesses of various values for our dimensions in order to present a guideline for researchers to choose the appropriate features (dimensions ’ values) to create or use an OL system for their own domain or application.

This Dissertation is dedicated to my father and mother--A. Baqui Khan, and Lutfe Ara Begum, who helped me grow into the person I am today and who have been "with me " in every sense in all phases of this Ph.D. journey. ii Acknowledgements All praises to almighty God, most gracious and most merciful, who allowed me to write this dissertation. He made me fortunate enough to be associated with so many competent and intellectually stimulating individuals who have guided me at every stage of my Ph.D. First, and foremost, I would like to thank Dennis McLeod, my research advisor and "mentor, " for his excellent guidance and encouragement throughout my research, and for his friendship during my years as a graduate student. His high standards of scholarship and intellectual integrity, as well as his openness and flexibility, contributed substantially to the understanding of the complex intersection of information retrieval, databases, and artificial intelligence which is presented in this dissertation. I was extremely fortunate to work under his supervision for my Ph.D.

This paper illustrated the application of cooccurrence theory to generate lightweight ontologies semi-automatically. First, the relationship of Information Retrieval (IR) and Artificial Intelligence (AI) is discussed in a general way. Then two case studies have been conducted to generate lightweight ontologies in specific domains (Information Retrieval domain and European part of CIA FactBook). Further discussion was articulated and future work was proposed, especially the possible future research direction on ontology learning. 1.

Designing ontologies is a key aspect of knowledge management and knowledge representation. We introduce a document based ontology building methodology for french legal documents. Unlike others methods, main knowledge will be here directly extracted from texts in two ways : (1) a rst knowledge acquisition from texts that extracts relevant terms of the domain and their relations and (2) a second acquisition that extract implicit knowledge from the documents. Our documents are the legal texts published every day in the Journal Ociel de la Republique francaise, the french ocial publication for legal texts. Our goal, designing an ontology from those texts, is to enable conceptual retrieval of the documents and to formalize the conceptual framework of an information system based on those documents. This ontology has clearly documentary purposes and is dedicated to normative texts. 1

Abstract not found

Taxonomies are an important resource for a variety of Natural Language Processing (NLP) applications. Despite this, the current stateof-the-art methods in taxonomy learning have disregarded word polysemy, in effect, developing taxonomies that conflate word senses. In this paper, we present an unsupervised method that builds a taxonomy of senses learned automatically from an unlabelled corpus. Our evaluation on two WordNet-derived taxonomies shows that the learned taxonomies capture a higher number of correct taxonomic relations compared to those produced by traditional distributional similarity approaches that merge senses by grouping the features of each word into a single vector. 1

In the past a lot of approaches concerning the integration of heterogeneous information sources are developed. In the last years the semantics, which play an important role during the integration task, come into the focus leading to the so called ontology-based integration approaches. This paper provides a survey of most prominent ontologybased integration approaches. The approaches are evaluated according four criterions, i.e. the role and the representation of the ontologies, the mapping relating sources and ontologies, and their support for ontology engineering. The evaluation gives an impression, how which problems are solved, and shows the need for further research.