Mobile social travellers guide
Mobile Social Travellers Guide is a recommender system whose domain is tourist attractions. The system utilizes the users social network which is Facebook and a photo sharing website which is Flickr to disambiguate between different location names referring to the same location and location names th...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Animo Repository
2010
|
| Subjects: | |
| Online Access: | https://animorepository.dlsu.edu.ph/etd_bachelors/11165 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | De La Salle University |
| Language: | English |
| Summary: | Mobile Social Travellers Guide is a recommender system whose domain is tourist attractions. The system utilizes the users social network which is Facebook and a photo sharing website which is Flickr to disambiguate between different location names referring to the same location and location names that are part-of another location. The data collection, disambiguation, and recommendation process are all hidden in the form of a social network game. The word relationships part-of and same-as are used for disambiguation.
The proponents were able to develop an online game for the web part and a mobile application for the mobile part of the system. The mobile application is an extension of the online game wherein the users are able to view the users account, interact with other users by being able to recommend locations to their customers and evaluate the recommendations given to them and receive recommendations while the users are on the go.
The disambiguation of location names were done through the online game and the mobile application. Both parts create and verify relationships, through the creation of Packages and Attractions, as well as with the user interaction. As a result, there were a total of 117 same relationships and a total of 234 part-of relationships that were created in the database. At the same time, the systems recall and precision for searching locations were 93.3 percent and 88.19 percent, respectively. |
|---|