DSKQ: A system for efficient processing of diversified spatial-keyword query

Publication Type:
Conference Proceeding
Citation:
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2017, 10538 LNCS pp. 280 - 284
Issue Date:
2017-01-01
Metrics:
Full metadata record
Files in This Item:
Filename Description Size
ADC_paper.pdfPublished version1.06 MB
Adobe PDF
© 2017, Springer International Publishing AG. With the rapid development of mobile portable devices and location positioning technologies, massive amount of geo-textual data are being generated by a huge number of web users on various social platforms, such as Facebook and Twitter. Meanwhile, spatial-textual objects that represent Point-of-interests (POIs, e.g., shops, cinema, hotel or restaurant) are increasing pervasively. Consequently, how to retrieve a set of objects that best matches the user’s submitted spatial keyword query (SKQ) has been intensively studied by the research communities and commercial organisations. Existing works only focus on returning the nearest matching objects, although we observe that many real-life applications are now using diversification to enhance the quality of the query results. Thus, existing methods fail to solve the problem of diversified SKQ efficiently. In this demonstration, we introduce DSKQ, a diversified in-memory spatial-keyword query system, which considers both the textual relevance and the spatial diversity of the results processing on road network. We present a prototype of DSKQ which provides users with an application-based interface to explore the diversified spatial-keyword query system.
Please use this identifier to cite or link to this item: