Four teams from the Department of Information Systems, Faculty of Computer Science and Information Technology (FCSIT), University of Malaya participated in the International Engineering Invention and Innovation Exhibition 2014 (I-ENVEX), which was held in Universiti Malaysia Perlis (UniMAP), from April 11 – 13, 2014. Young investors aged between 12 and 25 from all over the world participated in this exhibition.
The FCSIT team participated in the ICT, Multimedia, Telecommunications, Electricity & Electronic categories (Class E). The results and project descriptions are as follows:
One of the projects entitled “Personalized E-Learning System with the Adoption of a Help-Seeking Model”, focused on linking help-seeking cognitive skill to adaptive e-learning systems. The developed prototype is able to monitor the progress and information provided adaptively to students, and also recommends modules to assist the students. This project won the Gold Medal and also the overall Best Invention Award for the ICT category.
Two other projects were awarded the Silver Medals. The first project entitled “Document Retriever Using Users' Implicit Feedback” is a document retriever system that produces improved search results based on users’ implicit feedback. Feedback such as dwell time, text highlight, page review and click-through were gathered as users interact with the search engine, and they are then used to retrieve more relevant results in future. Comparisons against baseline techniques indicate that when these feedbacks were integrated, document retrieval precisions are improved. The second project entitled “An Efficient Rough Set-based Data Classification Software” was a new approach proposed based on rough set theory by taking into account the dependency of attributes of databases. The results also show the proposed technique to produce better performance compared to other baseline data classification techniques.
Finally, a Bronze Medal was awarded to the fourth project, entitled “Clustering of Kuala Lumpur Companies based on their co-Movement in the Stock Market”. This project focused on using Hadoop Map Reduce to store and process the KLSE big data, whereas k-means algorithm was used to perform the clustering, and finding the companies that had similar KLSE closing bids pattern to help the investors to predict a company’s next closing bid based on another company that have the similar trends.
