User centered evaluation of personalization in social web
2017-02-06T02:04:46Z (GMT) by
The emergence of Social Networking Systems (SNS) is one of the most noticeable drifts of the World Wide Web. Users leave a plethora of information over the web. This in return, results in increasing amount of data, and creates the problem of information overload. Personalized applications in social networking can address the problem of information overload by providing the user with most relevant data which is consistent with his/her preferences. However, personalization in SNS is still challenging, and many issues remain unresolved. Most of the work in SNS personalization is focused on predicted rating accuracy evaluation. Only few have gone towards the factors that affect personalization while the fundamental aim of personalization is to target user satisfaction level. A bottom up, user-centered approach is required in this regard. Consequently, a comprehensive view on user profiles would be significant in analyzing the user’s association with different kinds of social networks that could assist the designers of sophisticated recommender systems, to facilitate users with usage of their profiles data for personalized services. In practical aspects of personalization within SNS domain, the metrics of personalization adoption appears as a more significant and interesting problem than user modeling. The recent evidences of recurrent rollbacks of personalization features from SNS have raised questions regarding their acceptance and utilization. Moreover, the scarcity of the research on the factors determining adoption of personalized solutions in SNS environment highlighted a requirement to build up a theoretical understanding of the factors playing a role to influence adoption initiatives. Our research work focuses on understanding users’ intention to adopt personalization in SNS. In this order, thesis will follow two directions: (i) The identification of factors affecting the adoption of SNS personalization drawing on technology acceptance modeling; and (ii) The emergence of users’ personalization related trade-offs (e.g. trade-offs between perceived benefits vs. perceived cost associated with the use of SNS). We believe that the realization of these elements, will certainly lead to new assertions, and will help to focus on the topics that emerge as a prerequisite for personalization based on the social web profiles. The novelty of this research lies in (i) exploration and exploitation of a new model, the UTAUT2, to determine the acceptance of information technology innovation, and (ii) addition to the literature by integrating the privacy calculus with acceptance model and the contextual constructs of SNS. It is noteworthy that the new constructs of social influence, personal innovativeness, facilitating conditions, privacy concerns, and previous privacy invasion are well-suited with UTAUT2 constructs and are sited within the nomological organization of the baseline model. This strategy is expected to ensure a stable theory development and development of consistent model of drivers of end user acceptance in SNS context. This work also contributes to the community of practice, as the understanding of personalized services acceptance by disclosing personal information and their utilization is imperative to the SNS vendors. This class of academic studies is significant for linking MIS research to practice, by benefitting practitioners with the presentation of valuable strategies, for not only improving existing applications but also future developments.