A key challenge in mobile social media applications is how to present personalised content that is both geographically and temporally relevant. In this paper, we propose a new and generic temporal weighting function for improving location recommendations. First, we identify areas of interest to recommend by clustering geographic activity based on a trace of geotagged photos. Next, the clusters are temporally weighted using TF-IDF, in order to capture seasonality, and a decay scoring function to capture preference drift. Finally, these weights are combined with the cluster scores based on geographic relevance. We evaluate our recommender on a large dataset collected from Panoramio consisting of the top-100 most populated cities in the world and show that incorporating the proposed temporal weighting function improves recommendation quality.
%0 Conference Paper
%1 balbymarinho2012improving
%A Marinho, Leandro Balby
%A Nunes, Iury
%A Sandholm, Thomas
%A Santos, Caio
%A Ezequiel, Jordão
%A Pires, Carlos Eduardo
%B Proceedings of the 18th Brazilian symposium on Multimedia and the Web (WebMedia'12)
%C New York, NY, USA
%D 2012
%I ACM
%K geographic myown recommender temporal
%P 293-296
%T Improving Location Recommendations with Temporal Pattern Extraction
%U http://doi.acm.org/10.1145/2382636.2382698
%X A key challenge in mobile social media applications is how to present personalised content that is both geographically and temporally relevant. In this paper, we propose a new and generic temporal weighting function for improving location recommendations. First, we identify areas of interest to recommend by clustering geographic activity based on a trace of geotagged photos. Next, the clusters are temporally weighted using TF-IDF, in order to capture seasonality, and a decay scoring function to capture preference drift. Finally, these weights are combined with the cluster scores based on geographic relevance. We evaluate our recommender on a large dataset collected from Panoramio consisting of the top-100 most populated cities in the world and show that incorporating the proposed temporal weighting function improves recommendation quality.
%@ 978-1-4503-1706-1
@inproceedings{balbymarinho2012improving,
abstract = {A key challenge in mobile social media applications is how to present personalised content that is both geographically and temporally relevant. In this paper, we propose a new and generic temporal weighting function for improving location recommendations. First, we identify areas of interest to recommend by clustering geographic activity based on a trace of geotagged photos. Next, the clusters are temporally weighted using TF-IDF, in order to capture seasonality, and a decay scoring function to capture preference drift. Finally, these weights are combined with the cluster scores based on geographic relevance. We evaluate our recommender on a large dataset collected from Panoramio consisting of the top-100 most populated cities in the world and show that incorporating the proposed temporal weighting function improves recommendation quality.},
added-at = {2012-11-25T21:25:10.000+0100},
address = {New York, NY, USA},
author = {Marinho, Leandro Balby and Nunes, Iury and Sandholm, Thomas and Santos, Caio and Ezequiel, Jordão and Pires, Carlos Eduardo},
biburl = {https://www.bibsonomy.org/bibtex/24f6eb7891ca4e9b6356a249d8ede4b3f/lbalby},
booktitle = {Proceedings of the 18th Brazilian symposium on Multimedia and the Web (WebMedia'12)},
interhash = {3bcdf6667f59ac6dcdfd5c3b72fed221},
intrahash = {4f6eb7891ca4e9b6356a249d8ede4b3f},
isbn = {978-1-4503-1706-1},
keywords = {geographic myown recommender temporal},
pages = {293-296 },
publisher = {ACM},
timestamp = {2014-11-02T20:00:31.000+0100},
title = {Improving Location Recommendations with Temporal Pattern Extraction},
url = {http://doi.acm.org/10.1145/2382636.2382698},
year = 2012
}