Mohd Ridzwan Yaakub

Bio: Mohd Ridzwan Yaakub is an academic researcher from National University of Malaysia. The author has contributed to research in topics: Sentiment analysis & Feature selection. The author has an hindex of 9, co-authored 37 publications receiving 287 citations. Previous affiliations of Mohd Ridzwan Yaakub include Queensland University of Technology.

A Two-Sided Market Mechanisms Toward Designing a Big Data-Driven Business Model for Mobile Network Operators (MNOs)

Abstract: The mobile ecosystem has experienced a dramatic change as a result of a new entrance, new application, and new business model emerged. Recent trends have shown that the mobile industry is amidst a transformation toward platformisation of the key players' business model. This situation puts a traditional business model of mobile network operators (MNOs) under pressure as they failed to catch up due to lesser capabilities for providing a new value proposition and incentives for those sides of the market to achieve the two-sided platform compared to dominated parties. However, the paucity of extant literature has provided practical solutions to enhance MNOs business model leveraging contemporary information technology tools in combination with managerial design principles and concepts. This study investigates the application of Big Data –driven business model in order to transform MNOs from one-sided business model to two-sided business model. The design science research method (DSRM) will be used to build and evaluate the new business model. The proposed model will help the operators better serve their upstream customers and the end-users in order to counter the stiff competition existing in the mobile ecosystem, and contribute to the Big Data - driven business model.

A new method to discretize time to identify the milestones of online social networks

Abstract: Online social networks (OSNs) are complex time-varying networks due to the exponential growth in the number of users and the activities of those users. As the form of OSNs can change in each time frame, those working in domains such as community detection, event detection, big data analytics, recommender systems and marketing need to find a way to discretize time to identify the behavioural changes in the OSN over time. For dynamic domains, it is necessary to chunk the network into some time windows and monitor all these time windows. However, to date, many studies have only attempted to monitor a network using one-time window as one inseparable piece of information, which can lead to misinterpretation of the data. Existing methods predict the population growth of a network based on a whole growth rate, but a network has some distinct growth rates during its lifespan. Therefore, this study aims to propose a new method to discretize time to detect the milestones of OSNs. However, many parameters can affect OSN growth. Therefore, in this study, an OSN growth equation is formulated on the basis that the network follows a specific order and discipline in its growth. This study introduces a two-variable equation based on the number of users and the number of connections, which are two common variables in all OSNs, to identify behavioural changes in OSNs. Experiments conducted on six different datasets as well as on real Facebook and real Twitter data show that an OSN follows two different patterns during its lifespan. These two growth patterns differ markedly, and the point at which these two patterns meet is the milestone of the network.

A New Real-Time Link Prediction Method Based on User Community Changes in Online Social Networks

Abstract: The link prediction problem is becoming an important area of online social network (OSN) research. The existing methods that have been developed to address this problem mostly try to predict links based on structural information about the whole of the user lifespan. In addition, most of them do not consider user attributes such as user weight, density of interaction and geo-distance, all of which have an influence on the prediction of future links in OSNs due to the human-centric nature of these networks. Moreover, an OSN is a dynamic environment because users join and leave communities based on their interests over time. Therefore, it is necessary to predict links in real time. Therefore, the current study proposes a new method based on time and user attributes to predict links based on changes in user communities, where the changes in the user communities are indicative of users’ interests. The proposed method is tested on the UKM dataset and its performance is compared with that of 10 well-known methods and another community-based method. The area-under-the-curve results show that the proposed method is more accurate than all of the compared methods.

Data Analytics in Malaysian Education System: Revealing The Success of Sijil Pelajaran Malaysia From Ujian Aptitud Sekolah Rendah

Abstract: Information on the student’s cognitive abilities can help teachers to identify the strengths or potential of a student to plan a learning strategy. These data are collected through Ujian Aptitud Sekolah Rendah in year six (UASR), where the student’s potential can be detected five years earlier before they take their Sijil Pelajaran Malaysia (SPM). Unfortunately, these data have not yet used as a criterion in the student’s development plan. Therefore, this research has been done to see the strength of the connection between the student’s potential in UASR and their results in SPM. Through data visualization, a strong connection between a student’s potential in UASR and their succesfullness in SPM can be seen. 71.48% of students in the first cohort that have been proven to house high potential during thier year six in 2011 and 73.63% of the next cohort in 2012 have obtained a good SPM result that they took in 2016 and 2017 respectively. Further analisys using the decision tree technique shows a few other factors for the high potential student success in SPM. A few of them are their results in PT3, their grades in the Maths and Science subjects, academic stream, and the school’s achievement for that particular stream for the year before.

Assessment of Self-Identity Among Teens Towards Self-Crime Prevention Program

Abstract: Teenagers can easily expose to criminal activities through media social, surrounding, and peers. They tend to imitate some of these criminal activities such as cyber bullying, fighting, cyber grooming and cyber harassment without thinking of his or communities’ side-effect. Therefore, we develop a self-crime prevention modules encompassing faith, positivity, social relationship, role model and reflection tailored for school teens. Then, we make a list of survey questions using 4 Likert scale to measure their acceptance and adaptation after participating our self-crime module. About 105 school children aged 13-17 participated from Sekolah Menengah Titiwangsa and they responded positively with approximately 3.63 of average score.

I and i

Abstract: There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality. Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

Computational Thinking 計算論的思考

Abstract: It represents a universally applicable attitude and skill set everyone, not just computer scientists, would be eager to learn and use.

A review of unsupervised feature selection methods

Abstract: In recent years, unsupervised feature selection methods have raised considerable interest in many research areas; this is mainly due to their ability to identify and select relevant features without needing class label information. In this paper, we provide a comprehensive and structured review of the most relevant and recent unsupervised feature selection methods reported in the literature. We present a taxonomy of these methods and describe the main characteristics and the fundamental ideas they are based on. Additionally, we summarized the advantages and disadvantages of the general lines in which we have categorized the methods analyzed in this review. Moreover, an experimental comparison among the most representative methods of each approach is also presented. Finally, we discuss some important open challenges in this research area.

An up-to-date comparison of state-of-the-art classification algorithms

Abstract: Up-to-date report on the accuracy and efficiency of state-of-the-art classifiers.We compare the accuracy of 11 classification algorithms pairwise and groupwise.We examine separately the training, parameter-tuning, and testing time.GBDT and Random Forests yield highest accuracy, outperforming SVM.GBDT is the fastest in testing, Naive Bayes the fastest in training. Current benchmark reports of classification algorithms generally concern common classifiers and their variants but do not include many algorithms that have been introduced in recent years. Moreover, important properties such as the dependency on number of classes and features and CPU running time are typically not examined. In this paper, we carry out a comparative empirical study on both established classifiers and more recently proposed ones on 71 data sets originating from different domains, publicly available at UCI and KEEL repositories. The list of 11 algorithms studied includes Extreme Learning Machine (ELM), Sparse Representation based Classification (SRC), and Deep Learning (DL), which have not been thoroughly investigated in existing comparative studies. It is found that Stochastic Gradient Boosting Trees (GBDT) matches or exceeds the prediction performance of Support Vector Machines (SVM) and Random Forests (RF), while being the fastest algorithm in terms of prediction efficiency. ELM also yields good accuracy results, ranking in the top-5, alongside GBDT, RF, SVM, and C4.5 but this performance varies widely across all data sets. Unsurprisingly, top accuracy performers have average or slow training time efficiency. DL is the worst performer in terms of accuracy but second fastest in prediction efficiency. SRC shows good accuracy performance but it is the slowest classifier in both training and testing.

Mining Opinion Features in Customer Reviews.

Abstract: Now days, E-commerce systems have become extremely important. Large numbers of customers are choosing online shopping because of its convenience, reliability, and cost. Client generated information and especially item reviews are significant sources of data for consumers to make informed buy choices and for makers to keep track of customer’s opinions. It is difficult for customers to make purchasing decisions based on only pictures and short product descriptions. On the other hand, mining product reviews has become a hot research topic and prior researches are mostly based on pre-specified product features to analyse the opinions. Natural Language Processing (NLP) techniques such as NLTK for Python can be applied to raw customer reviews and keywords can be extracted. This paper presents a survey on the techniques used for designing software to mine opinion features in reviews. Elven IEEE papers are selected and a comparison is made between them. These papers are representative of the significant improvements in opinion mining in the past decade.