IJMECS Vol. 8, No. 2, Feb. 2016
Cover page and Table of Contents: PDF (size: 652KB)
In this study, we describe Information Systems students' decision making along their engagement with their final project regarding the complexity and innovations of their projects, and the technology they selected for the implementation. Data was gathered from projects' documentation; a questionnaire handed to the study participants, and from in-depth interviews conducted with representative group of them. Analysis of the data revealed that high achievers tend to develop innovative and complex final projects using major extensions of technologies learned in class while low achievers tend to develop simple and basic final projects using merely technologies learned in class or a minor extension of them. Surprisingly, some of the average and low achievers and none of the high achievers tended to use completely new technologies to gain relative advantage when applying for jobs, although this choice necessitated them to cope with large knowledge gaps.[...] Read more.
As an interdisciplinary group of people working on automatic sign language recognition, authors of the article developed concept how to facilitate the process of understanding Sign Language (SL) utterances by hearing learners. Concept is based on Polish SL, and was created in response to signals from the students who indicate understanding of SL messages as the hardest part of learning process, but can be easily adapted to other SLs. In comparison to speech corpora, SL databases constitute only a small fraction. From the other side thanks to the Internet there are available video recordings with native-signers, as well as tools which enable their analysis. Developed concept is based on using one of the available annotation tools used by professionals all over the world to describe SL corpora in SL classes. Making annotations helps students in understanding foreign language, corpora analysis helps to find objective rules governing SL, computer assisted language learning can be attractive way of study and Internet – great source of materials.[...] Read more.
Social network sites have become de factor in fostering human relationships and business prospects. Several social networks abound with little interoperability functionality that enables exchange of profiles of users. Though, proprietary Application Programming Interfaces (APIs) are provided as endpoints for applications in retrieval of user profile. Moreover, semantic web Friend of a Friend (FOAF) is now been used as a medium for realizing semantic social networks to be able to share user’s profile across sites. And since the goal of semantic web is to provide autonomous data centric system coupled on ontology and reasoning, we propose a novel communication protocol named iProc, and usable by autonomous agents that relies on the distributive nature of social network data in coalescing a virtually centralized social network and as well providing means to enlarge user’s connectivity to other users across different sites. This paper presents the architecture for a proposed iProc. Furthermore, an implementation of the FOAF files to be used was carried out and discussed.[...] Read more.
Image Processing, a subset of Computer Vision, is an important branch in modern technology. Edge detection is a subset of segmentation to detect object of interest. Different image edge detection filters and their evaluating parameters are introducing rapidly. But the performance of an edge detector is an open problem. In this paper different performance measures of edge detection have been discussed in details and their application on a hybrid filter using Bilateral and Canny is proposed. Its parametric performance has been evaluated and other well established or classical existing edge detecting filters have been compared with it to measure its efficiency.[...] Read more.
The use of Data mining techniques on medical data is dramatically soar for determining helpful things which are used in decision making and identification. The most extensive data mining techniques which are used in healthcare domain are, classification, clustering, regression, association rule mining, classification and regression tree (CART). The suitable use of data mining algorithm can enhance the quality of prediction, diagnosis and disease classification. Valuation of data mining techniques demand for medical data mining is the major goal here, particularly to examine the local frequent disease like heart ailments, breast cancer, lung cancer and so on. We examine for discovering the locally frequent patterns through data mining technique in terms of cost performance speed and accuracy.[...] Read more.
One natural and successful technique to have retrieved documents that is relevant to users’ intention is by expanding the original query with other words that best capture the goal of users. However, no matter the means implored on the clustered document while expanding the user queries, only a concept driven document clustering technique has better capacity to spawn results with conceptual relevance to the users’ goal. Therefore, this research extends the Concept Based Thesaurus Network document clustering techniques by using the Latent Semantic Analysis tool to identify the Best Fit Concept Based Document Cluster in the network. The Fuzzy Latent Semantic Query Expansion Model process achieved a better precision and recall rate values on experimentation and evaluations when compared with some existing information retrieval approaches.[...] Read more.
This study examines students’ overview of the students’ online capabilities of course that we have implemented in the MOODLE platform in a developing country and underlying information technology principles that are critical for an in-depth understanding of e-learning. A structured multiple choice questionnaire was distributed among students’ who were enrolled in the certificate of teaching in higher education course at the General Sir John Kotelawela Defence University, Sri Lanka. A total of 31 students participated in this study and completed written and online multiple choice questionnaire on MOODLE. The outcome of this study shows that there is a strong positive response on e-learning on MOODLE platform. Almost 61% of them were able to get extreme good results in the online examination and observed late submission in both printed and online examination. Although the outcome is preliminary in nature, the results provide cause for concern over the status of e-learning education in MOODLE platform in Sri Lanka which is highly satisfactory.[...] Read more.
Data Mining is a dominant tool for academic and educational field. Mining data in education atmosphere is called Educational Data Mining. Educational Data Mining is concerned with developing new methods to discover knowledge from educational/academic database and can be used for decision making in educational/academic systems. This work demonstrates an effective mining of students performance data in accordance with placement/recruitment process. The mining result predicts weather a student will be recruited or not based on academic and other performance during the entire course. To mine the students’ performance data, the data mining classification techniques such as – Decision tree- Random Tree and J48 classification models were built with 10 cross validation fold using WEKA. The constructed classification models are tested for predicting class label for new instances. The performance of the classification models used are tested and compared. Also the misclassification rates for the classification experiment are analyzed.[...] Read more.