IJMECS Vol. 11, No. 1, Jan. 2019
Cover page and Table of Contents: PDF (size: 230KB)
This study aims to examine how learners engaged in wiki-based collaborative writing when they were using their first language (L1) or second language (L2). Issues concerning similarities and differences in wiki collaborative writing activities, wiki participation levels, wiki interaction patterns and wiki collaboration levels between the L1 and L2 writing groups are discussed. This paper reports a case study of a group of Hong Kong secondary school students who were required to use “Google Sites” to complete their Liberal Studies group projects. Student’s wiki writings and comments in their L1 (Chinese) and L2 (English) were collected and examined. Results indicated that, while similarities in writing activity patterns, participation levels and collaboration levels were shown in the two groups, differences did exist in their interaction patterns, which were affected by the different ways they used the wiki comment sections. This study can help educators to become aware of the different needs of L1 and L2 groups and to implement wiki collaborative writing more effectively to support students.[...] Read more.
With the sudden growth of the internet and digital documents available on the web, the task of organizing text data has become a major problem. In recent times, text classification has become one of the main techniques for organizing text data. The idea behind text classification is to classify a given piece of text to a predefined class or category. In the present research work, SVM has been used with linear kernel using the One-V-Rest strategy. The SVM is trained using various data sets collected from various sources. It may so happen that some particular words were not so common around 5-6 years ago, but are currently prevalent due to recent trends. Similarly, new discoveries may result in the coinage of new words. This process can also be applied to text blogs which can be crawled and then analyzed. This technique should in theory be able to classify blogs, tweets or any other document with a significant amount of accuracy. In any text classification process, preprocessing phase takes the most amount of time – cleaning, stemming, lemmatization etc. Hence, the authors have used a multithreading approach to speed up the process. The authors further tried to improve the processing time of the algorithm using GPU parallelism using CUDA.[...] Read more.
Though University enrolment in Nigeria is on the increase, more males than females are still being enrolled today, which varies according to discipline as well as from one geopolitical zone of the country to another. This is more pronounced in Science, Technology, Engineering and Mathematics fields, as female enrolment seems to be higher in the commercial and arts courses than the sciences and engineering. Secondary data were obtained from the Nigerian Bureau of Statistics, based on the Joint Admission and Matriculation Board registrations for a period of 5 years, spanning 2011 to 2015. The data were classified based on the six geopolitical zones in Nigeria, and multivariate data analysis, supported by Multivariate Analysis of Variance was carried out on the data. The results obtained revealed that there is still a wide variance in male and female enrolment in these fields, with male enrolment being significantly higher than that of female candidates. It also revealed that female enrolment varies depending on the geopolitical zone, with female enrolment in Science, Technology, Engineering and Mathematics being generally higher in geographical zones in southern Nigeria compared with those in northern Nigeria. The results obtained were further compared with data obtained from previous researches and the comparison was discussed. In addition, this study offers recommendations on how to encourage more female participation in Science, Technology, Engineering, and Mathematics.[...] Read more.
Sentiment classification is the most rising research areas of sentiment analysis and text mining, especially with the massive amount of opinions available on social media. Recent results and efforts have demonstrated that there is no single strategy can mutually accomplish the best prediction performance on various datasets. There is a lack of existing researches to Arabic sentiment analysis compared to English sentiment analysis, because of the unique nature and difficulty of the Arabic language which leads to shortage in Arabic dataset used in sentiment analysis. An Arabic benchmark dataset is proposed in this paper for sentiment analysis showing the gathering methodology of the most recent tweets in different Arabic dialects. This dataset includes more than 151,000 different opinions in variant Arabic dialects which labeled into two balanced classes, namely, positive and negative. Different machine learning algorithms are applied on this dataset including the ridge regression which gives the highest accuracy of 99.90%.[...] Read more.
VANETs is the open model which stimulate in academia and industry oriented researches. However, the model is open and there are many violations in a communication of vehicle to vehicle (V2V) and Vehicle to Infrastructure (V2I). Any anonymous user may extract the useful information. Researchers have proposed many research proposal and solved issues related to VANET. The security is the major concern and to avoid mishappening in driving the vehicle. We proposed the authentication system that provides safety of the driver during travel on the roads. The proposed results deliver the following features: 1) Reliability of VANET model 2) Road Safety 3) Privacy of the vehicles 4) Authentication of message delivery to adjacent nodes. Finally, we provide a view point of how to detect the attacks and withdraw malicious node more efficiently.[...] Read more.
Clustering is one of the extensively used techniques in data mining to analyze a large dataset in order to discover useful and interesting patterns. It partitions a dataset into mutually disjoint groups of data in such a manner that the data points belonging to the same cluster are highly similar and those lying in different clusters are very dissimilar. Furthermore, among a large number of clustering algorithms, it becomes difficult for researchers to select a suitable clustering algorithm for their purpose. Keeping this in mind, this paper aims to perform a comparative analysis of various clustering algorithms such as k-means, expectation maximization, hierarchical clustering and make density-based clustering with respect to different parameters such as time taken to build a model, use of different dataset, size of dataset, normalized and un-normalized data in order to find the suitability of one over other.[...] Read more.