PowerPOI

Project Leader: Prof. Chia-Hui Chang

Team members: Hsiu-Min Chuang, Ting-Yao Kao, Chung-Ting Cheng, Ya-Yun Huang, Guo-Bin Chang,

Kai-Chien Yang, Chien-Fu Ling, Yuan-Hao Lin, Hung-Wei Chang

Abstract

With the popularity of mobile devices and smartphones, we have witnessed rapid growth in mobile applications and services, especially in location-based services (LBS). According to a mobile marketing survey, maps/location searches are among the most utilized services on smartphones. Points of interest (POIs), such as stores, shops, gas stations, parking lots, and bus stops, are particularly important for maps/location searches. Existing map services such as Google Maps and Wikimapia are constructed manually either professionally or with crowdsourcing. However, manual annotation is costly and limited in current POI search services. With the abundance of information on the Web, many business POIs can be extracted from the Web.

In this project, we focus on automatically constructing a POI database to enable business POI map searches. We propose techniques that are required to construct a POI database, including focused crawling, information extraction, and information retrieval techniques. We first crawl Yellow Page websites to obtain vocabularies of business names. These vocabularies are then investigated with search engines to obtain sentences containing these business names from search snippets in order to train a business-name recognition model. To extract POIs scattered across the Web, we propose a query-based crawler to find address-bearing pages that might be used to extract addresses and business names. We crawled 1.25 million distinct POI pairs scattered across the Web and implemented a POI search service via Apache Lucent's search platform, called Solr. The experimental results demonstrate that the proposed geographical information retrieval model outperforms Wikimapia and a commercial app called "What's the Number?"

Publication

Related Technologies (Provide the training datasets for download)

  • Address extraction: (Data for Chinese; Collection for English provided by Z. Yu)

  • Business-name recognition: (Data)

  • Address-POI name pairing: (Data, window size=100)

  • Address-POI name verification: (Data)

Acknowledgement

This project is partially sponsored by the Ministry of Science and Technology in Taiwan under grant MOST103-2221-E-008-094