ABSTRACT
The context awareness technology is derived from the context calculation; therefore, the “context” also inherits the advantage of “who, when, where, what” of pervasive computing. Recently, context information has been deemed as the foremost part of the Recommendation and push System (RS) (Feng, 2014). The traditional information push mined user information through the Internet and pushed the information after completing its analysis and processing. With the dramatic development of the IOT, letting the context information join the conventional two-dimensional push system “user-item” and building a three-dimensional push system based on the form of “user-item-context” has already become a mainstream trend of information push service. JinHai Feng proposed using indoor positioning technology to track user activity in the market, in line with the historical information that is recorded, which contains information about where the user came from and what commodities the user browsed to estimate the user’s hobbies and favorites, and then accurately recommending goods to the user that they might be interested in (Wang, 2012). LiCai Wang
1 INTRODUCTION
With the user at the core, the personalized information service is devoted to push more effective and pointed personalized information service to the user by studying some context information such as the user’s behavior, environment, and emotions. This makes it possible to obtain the user’s demand information automatically and decrease the user’s search decision. Personalized information service needs to provide for real time and individual demands of the users and fulfill their personalized needs, which means that the service should possess the characteristic of self-adaptability. With the conventional information push technology, the user demands are maintained stably, but lack individuation, real time, and renewability; therefore, the traditional methods of catching user message are no longer in use. With the boom of the Internet Of Things (IOT), any information that is linked to the user’s conduct, target, or the surrounding environment is all considered as an item that has some interactive relation with the user and the pervasive computing environment; thus, combining the context data with the user’s behavior would be a great way to figure out the user’s interest preference. Using the IOT technology to gain relative context information and considering a user’s behavior on
has been trying to judge whether the requirements of mobile users would suffer from the context by computing the volatility of mobile users’ behavior that is under constraint of the context, and to make sure of the extent of the influence by making use of the volatility (Gong, 2011). Combining mobile Internet and QRCode, XinWen Gong designed a mobile learning platform based on context awareness, in which the known context information of the users can be used as an important reference to provide the corresponding learning content (Li, 2012). Based on context awareness technology, XiaoYan Li established three application scenarios to provide shopping reminders for users automatically by combining user opinion with the context to mine users’ shopping decisions and underlying demands and accordingly sending shopping reminders to the users automatically (Adomavicius, 2005). Champiri and his coworkers hold the view that there are three conditions of context information: user context, file context, and environment background. They also emphasized combining the concept of context with user opinion to make academic recommendations in the academic field; therefore, a digital library recommendation system based on context awareness has been designed and implemented effectively (Baltrunas, 2011). Gavala and Kenteris used context information such as the current location, time, and condition to extend collaborative filtering technology and provide corresponding mobile travel guidance according to the user’s mobile device (Gavalas, 2011).
