An Approach for Service Discovery and Recommendation Using Contexts

Abstract

Given the large amount of existing Web services nowadays, it is time-consuming for users to find appropriate Web services to satisfy their diversity requirements. Context-aware techniques provide a promising way to help users obtain their desired services by automatically analyzing a user's context and recommending services for the user. Most existing context-aware techniques require system designers to manually define reactions to contexts based on context types (e.g., location) and context values (e.g., Toronto). Those context-aware techniques have limited support for dynamic adaptation to new context types and values. Due to the diversity of user's environments, the available context types and potential context values are changing overtime. It is challenging to anticipate a complete set of context types with various potential context values to provide corresponding reactions. In this chapter, we present an approach which analyzes dynamic changing context types and values, and formulates search criteria to discover desired services for users. More specifically, we use ontologies to enhance the meaning of a user's context values and automatically identify the relations among different context values. Based on the relations among context values, we infer the potential tasks that a user might be interested in, then recommend related services. A case study is conducted to evaluate the effectiveness of our approach. The results show that our approach can use contexts to automatically detect a user's requirements in given context scenarios and recommend desired services with high precision and recall.

A Context-Aware Intentional Service Prediction Mechanism in PIS

“…These two measures are defined in terms of a set of retrieved item and a set of relevant items. The precision represents how well a system retrieves only the relevant services, while the recall measures the ability of a system to retrieve all the relevant service [21]. The definition of recall and precision measures are defined as follows:…”