A Hybrid Dynamic Recommender System using Simple Additive Weighting Method for Service Search with Collaborative Filtering

This paper describes the development of a dynamic recommender system which provides clients the recommended services based on their preferences. This system is said to be dynamic since the weight of each criterion is defined by the user. In addition, preferences of other users is also included in ranking the recommended services through the use of collaborative filtering. This provides a specific user an idea of what others preferred and may help the user in finding the best offer. In this paper, a Simple Additive Weighting (SAW) method is used to create a function to calculate the ranking score of each possible service that match the user preferences. The criteria used for the service search includes the primary factors in choosing a service such as cost, location, and rating from other clients. In addition, one of the criteria is the preferences of other users which is processed through collaborative filtering. A web application is developed to provide interface for the clients where they can search services according to their preferences. In this paper, the service search is tested on event services. Usability testing is conducted and results revealed that users strongly agree with the usefulness of the system.