To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
A Hybrid Preference Oriented Collaborative Filtering Technique for Context-aware Point of Interest Search |
|
Author |
Hamid Turab Mirza, Naeem Qasim, Ibrar Hussain, Mujahid Rafiq, Safdar Ali, Hafiz Mahfooz Ul Haque |
| Citation |
Vol. 26 No. 1 pp. 71-84 |
|
Abstract |
Nowadays, location-based services are in high demand due to the ubiquitous use of mobile devices. In such services, the location of a user is commonly utilized as a search criterion to find user’s Point of Interest (PoI). Existing approaches suffer from various limitations such as ignoring user preferences in the search criteria, scalability issues along with data credibility problems in public evaluation strategies. Consequently, most users are not satisfied with the search results in the absence of such rich information. This work introduces a novel technique to search for K-nearest points which are preferable to the user, by utilizing searching time as well as query location. Specifically, this work has proposed a hybrid system that employs feedback learning algorithm, collaborative filtering and Google page rank algorithm. The feedback learning algorithm and collaborative filtering are used to enable continuous learning and improve the predictive accuracy respectively. Google page rank is utilized to increase the credibility of public evaluation while calculating the score of the PoI. The proposed system is experimentally evaluated on a benchmark data obtained from yelp.com. The results revealed a significant gain in performance and accuracy. |
|
Keywords |
Location-based search; preference learning; feedback learning; collaborative filtering; recommendation system. |
|
URL |
