Search in location based social network

• Main Author: Lew, Stephanie Yin Hui.
• Other Authors: School of Computer Engineering
• Format: Final Year Project
• Language: English
• Published: 2013
• Subjects: DRNTU::Engineering::Computer science and engineering::Information systems::Information storage and retrieval
• Online Access: http://hdl.handle.net/10356/55035
• Institution: Nanyang Technological University

With the advent of geo-positioning technologies, it is possible for a user to be able to check-in his or her location information online, especially to location based social networks such as Foursquare and Twitter. It is reported that web querying with local intent has also increased, especially originating from mobile users who are on the go. Thus, a search engine that can effectively capture location embedded information can come in useful in such situations. Such content can be conceptualized as spatial objects, which contain both spatial and textual information. With a location aware search engine, users can submit geographically constrained searches against a structured database of spatial objects. To improve the retrieval process of these spatial objects is the prime motivation of this project. A custom ranking score incorporating distance, textual relevancy and popularity of a location was designed in the process, to score and find top-k spatial objects. Four indexing schemes that could be used for the local spatial querying were discussed in the report. In particular, the Cartesian Tier plotting, Geohasing with prefixes and Sharding for scalability was expanded on and implemented with Lucene and ElasticSearch. Through the performance experiments conducted, it was found that the implementation of Geohashing with Lucene or ElasticSearch could support fast querying in a local spatial search system. A web application prototype was developed to visualize the results. The project can be potentially expanded to support other kinds of geospatial shapes such as polygons or lines, apart of venue points and also other types of spatial querying such as collective or relevant region querying that returns relevant groups of spatial objects.