ترکیبی از آزمایش های تجربی و تکنیک های مدل سازی : طراحی روش تحقیق برای برنامه های کاربردی شخصی تبلیغات تلفن همراه
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|2086||2008||15 صفحه PDF||سفارش دهید||9025 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Decision Support Systems, Volume 44, Issue 3, February 2008, Pages 710–724
We propose a design research approach combining behaviour and engineering techniques to better support user modeling in personalized mobile advertising applications. User modeling is a practical means of enabling personalization; one important method to construct user models is that of Bayesian networks. To identify the Bayesian network structure variables and the prior probabilities, empirical experimentation is adopted and context, content, and user preferences are the salient variables. User data collected from the survey are used to set the prior probabilities for the Bayesian network. Experimental evaluation of the system shows it is effective in improving user attitude toward mobile advertising.
With the development of wireless and mobile networks, mobile commerce (m-commerce) is creating significant benefits and new business opportunities for both mobile devices and services. According to Research and Markets, the number of mobile users worldwide is expected to climb to 2 billion by 2007, and the annual handset sales are predicted to generate more than U.S. $3 billion by 2009. The high penetration rate of mobile phones has resulted in the increasing use of handheld devices to deliver advertisements of products and services. Mobile advertising revenues, as indicated by many research firms, make up the largest share of m-commerce revenues. One may expect mobile advertising to be even more appealing to consumers who use location-sensitive and time-critical m-commerce applications. However, constraints in using both mobile networks and devices impose significant negative influences on the operational performance of m-commerce applications. For example, the small screens of the devices allow the user to view only limited pages of information . Many other inherent constraints exist in a mobile computing environment (e.g., minimal battery power on wireless devices , limited and error-prone wireless links  and , limited input and output capacity , the diversity of devices, and the myriad differences in user profiles), which should be taken into consideration in the design and development phases of m-commerce services and applications . The success of m-commerce largely depends on whether personalization can be well utilized to deliver highly personalized and context-sensitive (time- and location-dependent) information to mobile users. Providing personalized information to mobile users will create better customer satisfaction and will in turn increase the demand for mobile services.
نتیجه گیری انگلیسی
Design and behavioural science have complementary roles in the IS discipline. Lee  and Hevner et al.  observed that technology and behaviour are inseparable rather than dichotomous in an information system. The behavioural science concept aims at developing and verifying theories that explain or foresee individual or organizational behaviour. The design-science concept seeks to extend the boundaries of human and organizational capabilities by creating new and innovative objects. This paper has proposed a design research method that combines an empirical approach and modeling techniques for the design of a personalized mobile application. First, we conducted a questionnaire survey to verify the users' purchasing decision theory in the traditional marketing research and confirm that e-commerce settings can be applied to the m-commerce context. Further, we identified the factors that influence users' purchasing decisions and extend marketing theories in the m-commerce context in order to predict users' purchasing behaviour in the m-commerce environment. Next, those empirical findings were combined with the Bayesian-network-modeling technique for the development of a personalized mobile advertising environment. More concretely, those personalization component variables were used to build the Bayesian network structure, and the data collected from the empirical study were used to set the prior probabilities. A prototype implementing the user model, content model, context model, and a matching engine to deliver more relevant advertisements to the right mobile customers has been developed as a result.