Location Aware Keyword Query Suggestion Based on Document Proximity

Computer Department

The Keyword suggestions are the most basic feature of the search engine. Naive users don’t know how to express their queries correctly most of the time queries are short and unambiguous; keyword suggestion in web search assists users to access relevant information. Information will be more relevant if location of the users and user preferences are considered. The location aware keyword (LKPS) query suggestion method helps retrieve documents which relates to information provides by user, location where the user is located and the user profile defines the user interest. The user preference specific suggestion helps to improve the suggestions. The search results’ relevance is known to be correlated with the spatial proximity of the user and user preferences. The system proposes location-aware and user preferences specific keyword query suggestion framework. The framework introduces graph techniques and generate KD graph and keyword specific graph partitions. The partition algorithm is proposed to retrieve query suggestions. This algorithm follows Random walk with restart technique on candidate nodes in graph partitions and hence improves system efficiency. The results are conducted on AOL dataset of various sizes and evaluate the system performance.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.