TY - CONF AU - Doerfel, Stephan AU - Jäschke, Robert A2 - T1 - An Analysis of Tag-Recommender Evaluation Procedures T2 - Proceedings of the 7th ACM conference on Recommender systems PB - ACM CY - New York, NY, USA PY - 2013/ M2 - VL - IS - SP - 343 EP - 346 UR - http://doi.acm.org/10.1145/2507157.2507222 M3 - 10.1145/2507157.2507222 KW - 2013 KW - BibSonomy KW - core KW - evaluation KW - iteg KW - itegpub KW - l3s KW - myown KW - recsys KW - tag L1 - SN - 978-1-4503-2409-0 N1 - N1 - AB - Since the rise of collaborative tagging systems on the web, the tag recommendation task -- suggesting suitable tags to users of such systems while they add resources to their collection -- has been tackled. However, the (offline) evaluation of tag recommendation algorithms usually suffers from difficulties like the sparseness of the data or the cold start problem for new resources or users. Previous studies therefore often used so-called post-cores (specific subsets of the original datasets) for their experiments. In this paper, we conduct a large-scale experiment in which we analyze different tag recommendation algorithms on different cores of three real-world datasets. We show, that a recommender's performance depends on the particular core and explore correlations between performances on different cores. ER - TY - CONF AU - Doerfel, Stephan AU - Jäschke, Robert A2 - T1 - An analysis of tag-recommender evaluation procedures T2 - Proceedings of the 7th ACM conference on Recommender systems PB - ACM CY - New York, NY, USA PY - 2013/ M2 - VL - IS - SP - 343 EP - 346 UR - https://www.kde.cs.uni-kassel.de/pub/pdf/doerfel2013analysis.pdf M3 - 10.1145/2507157.2507222 KW - 2013 KW - bibsonomy KW - bookmarking KW - collaborative KW - core KW - evaluation KW - folkrank KW - folksonomy KW - graph KW - iteg KW - itegpub KW - l3s KW - recommender KW - social KW - tagging L1 - SN - 978-1-4503-2409-0 N1 - N1 - AB - Since the rise of collaborative tagging systems on the web, the tag recommendation task -- suggesting suitable tags to users of such systems while they add resources to their collection -- has been tackled. However, the (offline) evaluation of tag recommendation algorithms usually suffers from difficulties like the sparseness of the data or the cold start problem for new resources or users. Previous studies therefore often used so-called post-cores (specific subsets of the original datasets) for their experiments. In this paper, we conduct a large-scale experiment in which we analyze different tag recommendation algorithms on different cores of three real-world datasets. We show, that a recommender's performance depends on the particular core and explore correlations between performances on different cores. ER - TY - CONF AU - Mitzlaff, Folke AU - Atzmüller, Martin AU - Benz, Dominik AU - Hotho, Andreas AU - Stumme, Gerd A2 - T1 - Community Assessment using Evidence Networks T2 - Proceedings of the Workshop on Mining Ubiquitous and Social Environments (MUSE2010) PB - CY - Barcelona, Spain PY - 2010/ M2 - VL - IS - SP - EP - UR - http://www.kde.cs.uni-kassel.de/ws/muse2010 M3 - KW - 2010 KW - assessment KW - bibsonomy KW - community KW - evaluation KW - evidence KW - itegpub KW - l3s KW - myown KW - networks L1 - SN - N1 - N1 - AB - Community mining is a prominent approach for identifying (user) communities in social and ubiquitous contexts. While there are a variety of methods for community mining and detection, the effective evaluation and validation of the mined communities is usually non-trivial. Often there is no evaluation data at hand in order to validate the discovered groups. This paper proposes evidence networks using implicit information for the evaluation of communities. The presented evaluation approach is based on the idea of reconstructing existing social structures for the assessment and evaluation of a given clustering. We analyze and compare the presented evidence networks using user data from the real-world social

bookmarking application BibSonomy. The results indicate that the evidence

networks reflect the relative rating of the explicit ones very well. ER - TY - CONF AU - Mitzlaff, Folke AU - Atzmüller, Martin AU - Benz, Dominik AU - Hotho, Andreas AU - Stumme, Gerd A2 - T1 - Community Assessment using Evidence Networks T2 - Proceedings of the Workshop on Mining Ubiquitous and Social Environments (MUSE2010) PB - CY - Barcelona, Spain PY - 2010/ M2 - VL - IS - SP - EP - UR - http://www.kde.cs.uni-kassel.de/ws/muse2010 M3 - KW - 2010 KW - assessment KW - bibsonomy KW - community KW - evaluation KW - evidence KW - itegpub KW - l3s KW - myown KW - networks L1 - SN - N1 - N1 - AB - Community mining is a prominent approach for identifying (user) communities in social and ubiquitous contexts. While there are a variety of methods for community mining and detection, the effective evaluation and validation of the mined communities is usually non-trivial. Often there is no evaluation data at hand in order to validate the discovered groups. This paper proposes evidence networks using implicit information for the evaluation of communities. The presented evaluation approach is based on the idea of reconstructing existing social structures for the assessment and evaluation of a given clustering. We analyze and compare the presented evidence networks using user data from the real-world social

bookmarking application BibSonomy. The results indicate that the evidence

networks reflect the relative rating of the explicit ones very well. ER -