Demo: Using Personalized PageRank for Keyword Based Sensor Retrieval

The paper proposes a system for advanced sensor retrieval. Sensors provide a large volume of data about the environment. The abundance and variety of the sensor data has given rise to increased interest in the development of applications involving sensor networks. The large volumes of data produced by such sensor networks require special techniques not only for management and processing but also for its discovery. We propose and implement a system for sensor search, based on matching user’s given keywords against information extracted from standardized sensor descriptions. For ranking the results we use the Personalized PageRank algorithm and apply filtering based on geo-location. Illustrative examples are presented to show how a user can interact with our system and what the benefits of ranking the search results are.