Recommender system for selection of touristic advisors

Andrej Levičnik (2013) Recommender system for selection of touristic advisors. EngD thesis.

Preview

Abstract

We created a prototype recommender system, advising tourists on tourist guide choice. We examined the area of recommender systems and described two typical approaches to recommender systems design. Our recommender system profiles users based on their activity on social network Facebook. We developed an algorithm for retreival of textual data from Facebook and an algorithm for its exploitation via language technologies. We also used a function for translation of text into vector form using TF-IDF schema and cosine similarity function. The system was evaluated using manual recommendations. The results show a strong correlation between recommended guides and manually picked choices.

Item Type:

Thesis (EngD thesis)

Keywords:

recommender systems, Facebook, natural language processing tools, TF-IDF, cosine similarity

Number of Pages:

Language of Content:

Slovenian

Mentor / Comentors:

Name and Surname	ID	Function
izr. prof. dr. Marko Robnik Šikonja	276	Mentor

Link to COBISS:

http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=10202964)

Institution:

University of Ljubljana

Department:

Faculty of Computer and Information Science

Item ID:

2223

Date Deposited:

26 Sep 2013 11:05

Last Modified:

18 Oct 2013 13:07

URI:

http://eprints.fri.uni-lj.si/id/eprint/2223

Actions (login required)

View Item