Recommendation-Assisted Personal Web

With the growing establishment of Internet infrastructure, more and more online services become available to end user, which in turn promotes the prosperity of the Internet. However, two issues emerge during this information increase. First, users have to access many individual sites to get their services, which consumes lots of time and contains some duplicate work. Second, user traces in different websites could have been used to provide more personalized services. Given these observations, this position paper proposes a recommendation-assisted personal web system based on existing work on personal web and recommendation systems. This system can integrate several web services to form a personal web, derive request-specific user data, and provide a personalized service by content based filtering and user intention inference. Using a research assistant application as a case study, we show how this framework helps to deliver personalized services.