OPINION MINING ON EDUCATIONAL MOBILE APPLICATIONS
Institute for Information Industry (TAIWAN)
About this paper:
Appears in:
INTED2014 Proceedings
Publication year: 2014
Pages: 561-565
ISBN: 978-84-616-8412-0
ISSN: 2340-1079
Conference name: 8th International Technology, Education and Development Conference
Dates: 10-12 March, 2014
Location: Valencia, Spain
Abstract:
Emerging technologies, such as multimedia, the Internet, mobile devices, etc have derived versatile learning methodologies including game-based learning, web-based learning and mobile learning. According to a forecast made by Ambient Insight, the worldwide market for game-based learning and mobile learning will reach $2.5 billion and $9.1 billion by 2015 respectively, while mobile educational applications have a big market share in both game-based learning and mobile learning marketplaces.
As of February 2011, there are more than two hundred app stores around the world. Among them, Amazon Appstore is a global mass-market store where customers can browse and download various categories of mobile applications as well as sharing experience by rating and posting reviews. There are more than ten thousand educational apps on Amazon Appstore, most of which are free or under two dollars. To make significant profit, customer satisfaction and positive reviews are important for mobile applications to trigger e-words of mouth effect or in-app purchases. With opinion mining technologies, user reviews can be analyzed to uncover critical product features which influence customer attitudes.
In this study, we first collected reviews from both popular educational mobile applications and other low rated ones on Amazon Appstore. Then, we applied opinion mining on these reviews to find which features are most concerned by users. The results show that in reviews of popular mobile applications, user reviews focus on game play and functionality of the applications. On the other hand, in reviews of the low rated applications, users argued the pricing and in-app purchase most. To sum it up, this work aids mobile application developers in finding the critical features that receive very favorable responses from users.Keywords:
Opinion Mining, Mobile learning.