Word Sense Disambiguation for Ontology Learning

2013 | Americas Conference on Information Systems | Citations: 0

Authors: Wimmer, Hayden; Zhou, Lina

Abstract: Ontology learning aims to automatically extract ontological concepts and relatio ... Expand

Abstract: Ontology learning aims to automatically extract ontological concepts and relationships from related text repositories and is expected to be more efficient and scalable than manual ontology development. One of the challenging issues associated with ontology learning is word sense disambiguation (WSD). Most WSD research employs resources such as WordNet, text corpora, or a hybrid approach. Motivated by the large volume and richness of user-generated content in social media, this research explores the role of social media in ontology learning. Specifically, our approach exploits social media as a dynamic context rich data source for WSD. This paper presents a method and preliminary evidence for the efficacy of our proposed method for WSD. The research is in progress toward conducting a formal evaluation of the social media based method for WSD, and plans to incorporate the WSD routine into an ontology learning system in the future. Collapse

Topics: knowledge representation social media Twitter Java application programming interface

Methods: ontological modelling design artifact WordNet word sense disambiguation computational algorithm

Who is talking? An ontology-based opinion leader identification framework for word-of-mouth marketing in online social blogs

2011 | Decision Support Systems | Citations: 0

Authors: Li, Feng; Du, Timon C.

Abstract: Online social blogs have gained popularity recently. They provide an effective c ... Expand

Abstract: Online social blogs have gained popularity recently. They provide an effective channel for word-of-mouth (WoM) marketing to promote products or service. In WoM marketing, an opinion leader, who is normally more interconnected and has a higher social standing, can deliver product information, provide recommendations, give personal comments, and supplement professional knowledge that help companies to promote their products. Many theories have been put forward about social networks, but few address the issue of opinion leader identification. This study proposes a framework to identify opinion leaders using the information retrieved from blog content, authors, readers, and their relationships, which we call BARR for short. We first build ontology for a marketing product and then collect parameters from BARR to identify “hot topics” related to the product. These hot topics are then associated with information disseminators, or opinion leaders. Marketers can use BARR to track blogs written by opinion leaders and identify their opinions to form effective marketing strategies. Collapse

Topics: blog knowledge representation website Apple iPhone web page

Methods: ontological modelling TOPSIS trade-off analysis method domain ontology ontology extraction

Managing knowledge on the Web – Extracting ontology from HTML Web

2009 | Decision Support Systems | Citations: 0

Authors: Du, Timon C.; Li, Feng; King, Irwin

Abstract: In recent years, the Internet has become one of the most important sources of in ... Expand

Abstract: In recent years, the Internet has become one of the most important sources of information, and it is now imperative that companies are able to collect, retrieve, process, and manage information from the Web. However, due to the sheer amount of information available, browsing web content by searches using keywords is inefficient, largely because unstructured HTML web pages are written for human comprehension and not for direct machine processing. For the same reason, the degree of web automation is limited. It is recognized that semantics can enhance web automation, but it will take an indefinite amount of effort to convert the current HTML Web into the Semantic Web. This study proposes a novel ontology extractor, called OntoSpider, for extracting ontology from the HTML Web. The contribution of this work is the design and implementation of a six-phase process that includes the preparation, transformation, clustering, recognition, refinement, and revision for extracting ontology from unstructured HTML pages. The extracted ontology provides structured and relevant information for applications such as e-commerce and knowledge management that can be compared and analyzed more effectively. We give detailed information on the system and provide a series of experimental results that validate the system design and illustrate the effectiveness of OntoSpider. Collapse

Topics: web page knowledge representation website Java web application

Methods: ontological modelling cluster analysis machine learning ontology extraction experiment

Automatic Domain Ontology Extraction for Context-Sensitive Opinion Mining

2009 | International Conference on Information Systems | Citations: 0

Authors: Lau, Raymond Y K; Lai, Chapmann C L; Ma, Jian; Li, Yuefeng

Abstract: Automated analysis of the sentiments presented in online consumer feedbacks can ... Expand

Abstract: Automated analysis of the sentiments presented in online consumer feedbacks can facilitate both organizations’ business strategy development and individual consumers’ comparison shopping. Nevertheless, existing opinion mining methods either adopt a context-free sentiment classification approach or rely on a large number of manually annotated training examples to perform contextsensitive sentiment classification. Guided by the design science research methodology, we illustrate the design, development, and evaluation of a novel fuzzy domain ontology based contextsensitive opinion mining system. Our novel ontology extraction mechanism underpinned by a variant of Kullback-Leibler divergence can automatically acquire contextual sentiment knowledge across various product domains to improve the sentiment analysis processes. Evaluated based on a benchmark dataset and real consumer reviews collected from Amazon.com, our system shows remarkable performance improvement over the context-free baseline. Collapse

Topics: knowledge representation electronic commerce laptop computer application programming interface information retrieval

Methods: sentiment analysis domain ontology machine learning design artifact ontological modelling

Theories: fuzzy set theory function-behaviour-structure model

Information Granulation for the Design of Granular Information Retrieval Systems

2008 | International Conference on Information Systems | Citations: 2

Authors: Lau, Raymond Y K; Lai, Chapmann C L

Abstract: With the explosive growth of the amount of information stored on computer networ ... Expand

Abstract: With the explosive growth of the amount of information stored on computer networks such as the Internet, it is increasingly more difficult for information seekers to retrieve relevant information. Traditional document ranking functions employed by Internet search engines can be enhanced to improve the effectiveness of information retrieval (IR). This paper illustrates the design and development of a granular IR system to facilitate domain specific search. In particular, a novel computational model is designed to rank documents according the searchers’ specific granularity requirements. The initial experiments confirm that our granular IR system outperforms a classical vector-based IR system. In addition, user-based evaluations also demonstrate that our granular IR system is effective when compared with a well-known Internet search engine. Our research work opens the door to the design and development of the next generation of Internet search engines to alleviate the problem of information overload. Collapse

Topics: search engine knowledge representation internet technology Google+ information overload

Methods: experiment domain ontology ontological modelling computational algorithm descriptive statistic