Jer Lang Hong

Exploring students' perceptions of learning management system: An empirical study based on TAM

This research aims to explore the perceptions of students in using learning management system (LMS) in order to enhance the learning process of students. The research also identifies issues that students faced when using LMS in order to improve the ease of use of LMS. Technology Acceptance Model (TAM) is used as the theoretical framework to explain the adoption of LMS in term of perceived ease of use and perceived usefulness. Respondents agree that LMS is easy to use and useful to them as a repository. However, results indicate that there is a lack of interaction among lecturers and peers.

Ontological based webpage segmentation

In World Wide Web, images come together with valuable surrounding contextual information such as texts. There are very few efforts performed to address problems in mining of the images with their surrounding contextual information; although this information has long been extracted for wide range of uses such as image annotation, clustering of images, inference of image semantic content and image indexing. In this proposal, a genuine ontology based webpage segmentation algorithm is developed for extracting web images with its associated contextual information according to its semantic characteristics (e.g. synonymous words) as they appear on webpages.

OntoExtract - Automated extraction of records using multiple ontologies

Current search engines require an accurate yet fast automated extractor to extract relevant information from deep web for the users. Human users usually enter search queries and the search engines will then locate the desire information of interest by disambiguate the search query accordingly. The queries will then be passed on to multiple search engines for further processing. These search engines will then return the search results to the main search engine. However, data returned from these search engines are usually varied and presented in numerous formats and layouts. To extract them, we need automated extractor to filter out irrelevant information and locate the correct information. Current trends focused on using ontologies to automatically extract this information with high accuracy. To the best of our knowledge, no works have been made on using multiple ontologies (using many ontology techniques) to automatically extract information from deep webs. In this paper, we demonstrate that multiple ontologies technique can achieve higher accuracy when extracting data from the deep web.

Region based data extraction

Wrappers are tools used to extract relevant information from HTML pages. Current approaches use DOM tree, visual cue, and ontology to extract data. DOM tree based techniques are fast and simple. However, they are not as accurate as visual based wrappers due to lack of additional information needed to perform data extraction. Visual based wrappers, on the other hand, are slow due to the extra processing needed to obtain visual cue from the underlying browser rendering engine. Ontology based wrappers are accurate, but they are also slow and need manual tuning to operate them. In this paper, we propose a novel visual based wrapper to extract information from HTML pages. Our wrapper uses visual cue to eliminate unnecessary regions, hence reduces the running time of extraction task as our wrapper only needs to consider the relevant region for extraction. Then, our wrapper removes irrelevant data from the relevant region using visual cue. Experiment results show that our wrapper outperforms state-of-the-art wrapper WISH in data extraction.

Contextual information for image retrieval systems

Images in web come with valuable contextual information, which is useful for indexing and retrieval systems. However, there is no current acceptable standard which define the boundary and region where this information is located. In this paper, we discuss the state of the art image retrieval systems and address the current problem with these systems. We then formulate a hypothesis that contextual information can be found within the region nearby the images and we conduct a user study to validate our approach.

ELITE - A Novel Ranking Algorithm for Social Networking Sites Using Generic Scoring Function

Social networks need to manage and control the drift of insane amount of information by filtering and ranking everything in order to ensure they are right there for users’ viewing pleasure. However, the realization of social networks ranking is currently dictated by fairly straightforward optimization algorithms. Hence, there is a need for a newly enhanced and improved ranking algorithm to be formulated since users have been occasionally seeing what they should not. A composition of a generic score and a collective score that would equate to a whopping new-fangled algorithm called E.L.I.T.E. which comprises of five essential elements Engagement-U, Lifetime, Impression, Timeframe and Engagement-O in ensuring a more accurate result for users to see more of what they care about, less of what they do not and more of who they are interested in, less of who they are not. Engagement-U is the affinity between users measured by the relationships and other related interests between them, Lifetime is a trace of users’ past based on their positive, neutral and even negative interactions and actions with other users, Impression is the weight of each object determined by the number of positive responses from users, Timeframe is the timeline scoring technique in which an object naturally loses its value as time passes and Engagement-O is the attraction of users to objects measured between objects and associated interests of users.

MalayWordNet — A novel lexical database for Malay language

World Wide Web contains huge amount of data available in different languages across the world. Web browsers are tools used to display the data in graphical forms. With the evolution of Web 3.0, data has become an important part of human daily tasks, where it is used to process information, and formulate important decision rules for many organizations. Current tools used to conceptualize data are catered for some of the world well known languages such as English. However, these tools may not be able to support other languages as there are a wide range of languages with different syntax and representation. In this paper, we present a novel lexical semantic based database tool called MalayWordNet, specifically written for Malay language. Our tool is helpful for high end semantic based applications which use Malay language as part of their data presentation.

A novel framework for validating dynamic web applications

World Wide Web contains an abundant amount of websites which are deemed invalid as they do not adhere to the HTML standards defined under W3C. Websites which are not well-formed may cause performance degradation for some applications and poses compatibility issues with some browsers. Furthermore, it is harder to render a non well formed webpage using state of the art web browsers. Therefore, a wide spectrum of HTML document validation tools are developed to validate HTML pages. However, these tools are able to validate only static based HTML pages, which is a drawback as it is not catered for current state of the art dynamic server side generated pages. To resolve this issue, we propose a novel framework for validating dynamic web applications. In our framework, we develop set of heuristic rules considering the various languages syntax and operation for well known server side languages such as PHP, Java EE, and ASP.NET. Our proposed framework is able to robustly validate dynamic based HTML pages with high efficiency.

Integration of heterogeneous data for real world domain

There are a large number of companies that have abundant amount of data presented in unstructured forms. For example, accounting firms and banking sectors have clients who may provide financial data for them. However, these clients have data which is of diverse nature. To make this data presentable in a meaningful way, companies usually use their employees to key in data manually to their own systems. This approach is tedious and labor intensive. To facilitate the data processing, these companies normally prefer to have automated softwares that can easily recognize the different formats of data, process and present them in a usable form suitable for further processing. However, there are currently no automated software tools which can perform this task. In this paper, we propose a novel ontological tool in this study to extract data from heterogeneous sources. Our tool will be able to analyze the semantic properties of the various data.

OntoLabel - data labeling for deep web using WordNet

Wrappers are used to extract relevant information from the deep web, align, tabulate, and label these data for the users to recognize the contents simply and quickly. Existing wrappers use DOM Tree and visual cue to label data. These wrappers use several sets of heuristic rules to determine the label for a particular data, which may not be applicable to certain groups of data with words which are nearly similar in meaning. In this paper, we propose an ontological wrapper for labeling data using existing lexical database for English, WordNet. Our wrapper could label a wide range of data records without using any specific assumptions for the data structure. Instead of examining data structure and layout, our wrapper assigns label using the contents of the data. Experimental results show that our wrapper could label data records with high accuracy.