%0 Book Section %1 ma2013tagrank %A Ma, Feichao %A Wang, Wenqing %A Deng, Zhihong %B Social Media Retrieval and Mining %C Berlin/Heidelberg %D 2013 %E Zhou, Shuigeng %E Wu, Zhiang %I Springer %K bookmarking collaborative core folksonomy recommender social tagging tagrank %P 80--91 %R 10.1007/978-3-642-41629-3_7 %T TagRank: A New Tag Recommendation Algorithm and Recommender Enhancement with Data Fusion Techniques %U http://dx.doi.org/10.1007/978-3-642-41629-3_7 %V 387 %X In the era of web2.0, more and more web sites, such as Lastfm, Delicious and Movielens, provide social tagging service to help users annotate their music, urls and movies etc. With the help of tags, users can organize and share their online resources more efficiently. In this paper, we propose a new tag recommendation algorithm TagRank which is based on random walk model. We also explore three data fusion techniques to make more powerful hybrid tag recommenders using TagRank, two collaborative filtering based algorithms and three tag popularity based algorithms. In order to find appropriate individual recommenders to make hybrid, we propose a greedy selection algorithm. We test and verify our proposed TagRank and greedy selection algorithm on three real-world datasets and experimental results show that our methods are efficient in terms of precision, recall and F1. %@ 978-3-642-41628-6

@incollection{ma2013tagrank, abstract = {In the era of web2.0, more and more web sites, such as Lastfm, Delicious and Movielens, provide social tagging service to help users annotate their music, urls and movies etc. With the help of tags, users can organize and share their online resources more efficiently. In this paper, we propose a new tag recommendation algorithm TagRank which is based on random walk model. We also explore three data fusion techniques to make more powerful hybrid tag recommenders using TagRank, two collaborative filtering based algorithms and three tag popularity based algorithms. In order to find appropriate individual recommenders to make hybrid, we propose a greedy selection algorithm. We test and verify our proposed TagRank and greedy selection algorithm on three real-world datasets and experimental results show that our methods are efficient in terms of precision, recall and F1.}, added-at = {2013-12-10T15:51:41.000+0100}, address = {Berlin/Heidelberg}, author = {Ma, Feichao and Wang, Wenqing and Deng, Zhihong}, biburl = {https://www.bibsonomy.org/bibtex/2ba7d071a30dddb474e81d112f2b28b65/jaeschke}, booktitle = {Social Media Retrieval and Mining}, doi = {10.1007/978-3-642-41629-3_7}, editor = {Zhou, Shuigeng and Wu, Zhiang}, interhash = {c2dba3863f3f993d1ae777058d0af9b6}, intrahash = {ba7d071a30dddb474e81d112f2b28b65}, isbn = {978-3-642-41628-6}, keywords = {bookmarking collaborative core folksonomy recommender social tagging tagrank}, language = {English}, pages = {80--91}, publisher = {Springer}, series = {Communications in Computer and Information Science}, timestamp = {2014-07-28T15:57:31.000+0200}, title = {TagRank: A New Tag Recommendation Algorithm and Recommender Enhancement with Data Fusion Techniques}, url = {http://dx.doi.org/10.1007/978-3-642-41629-3_7}, volume = 387, year = 2013 }