Information Retrieval in Folksonomies: Search and Ranking

By Andreas Hotho, Robert Jäschke, Christoph Schmitz, Gerd Stumme. Extended Semantic Web Conference, 2006

1,063 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/9f1b0fa1c157a1255af4a10c434da73a0f881221

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Lars Schmidt-Thieme
2. Steffen Rendle
3. Sophie Ahrens
4. Qiang Yang
5. Xiangnan He
6. David F. Gleich
7. Yong Yu
8. Alexandros Nanopoulos

In Information Systems

1. Sophie Ahrens
2. Steffen Rendle
3. Lars Schmidt-Thieme
4. Xiangnan He
5. Bamshad Mobasher
6. Jie Lu
7. Yong Yu
8. Yongfeng Zhang

In Statistical and Nonlinear Physics

1. Sophie Ahrens
2. David F. Gleich
3. Lars Schmidt-Thieme
4. Steffen Rendle
5. Alexandros Nanopoulos
6. Filippo Menczer
7. Ciro Cattuto
8. Jie Lu

Citations in the corpus

• Recommender Systems (2012, methodology): “The FolkRank [272], a modified PageRank algorithm, was proposed to rank tags in folksonomies by assuming that important tags are given by important users.”
• Collaborative Filtering beyond the User-Item Matrix (2014, neutral): “In addition to exploiting tags for recommending items, research topic [Hotho et al. 2006; J¨aschke et al. 2008], which, however, falls outside the scope of this survey.”
• Pairwise interaction tensor factorization for personalized tag recommendation (2010, methodology): “Furthermore, we compare to other recent tag recommender methods: HOSVD [22], FolkRank and Adapted Pagerank [5] as well as the upper bound for non-personalized tag recom­menders [18].”
• Adnostic: Privacy Preserving Targeted Advertising (2010, background): “Folksonomies are an active area of research in the semantic web, information retrieval and data-mining communities [37, 23], and therefore we are making our dataset available to other researchers at http://crypto.”
• Tag-Aware Personalized Recommendation Using a Hybrid Deep Model (2017, background): “0, social tagging systems are introduced by many websites, where users can freely annotate online items using arbitrary tags (commonly known as folksonomy [Hotho et al., 2006]).”
• Tag-Aware Personalized Recommendation Using a Deep-Semantic Similarity Model with Negative Sampling (2016, background): “In the era of Web 2.0, social tagging systems are introduced by many websites, where users can freely annotate online items using arbitrary tags (commonly known as folk-sonomy [5]).”
• Learning optimal ranking with tensor factorization for tag recommendation (2009, methodology): “The best results reported in terms of precision and recall, were given by the FolkRank algorithm [3], an adaptation of the well known PageRank.”
• Optimizing web search using social annotations (2007, methodology): “proposed Adapted PageRank and FolkRank to find communities within the folksonomy but have not applied them to web search [1].”
• Recommender system application developments: A survey (2015, methodology): “Tag recommender system e-resource CF Web-based Individual 2011 [124] FolkRank e-resource CF, CB Web-based Individual 2006-20 07 [125, 126]”
• Personalized recommendation in social tagging systems using hierarchical clustering (2008, methodology): “In [9], a novel algorithm, FolkRank, for search and ranking in folksonomies is proposed that dependson interrelated tags, resources and users.”
• Tag recommendations based on tensor dimensionality reduction (2008, methodology): “We use Folkrank [8] and the Collaborative Tag Suggestions [18] al­gorithm (known as Penalty-Reward algorithm), denoted as FolkRank and PR, respectively as comparisons to our ap­proach.”
• Solving the apparent diversity-accuracy dilemma of recommender systems (2008, neutral): “Topical information and other measures of item-item similarity can be used to further diversify recommendation lists [24]: usergenerated classifications such as tags [33, 34, 35] may be useful here.”
• On social networks and collaborative recommendation (2009, background): “What is more, the act of collaborative annotation establishes a folksonomy [10] of tracks, that is a user-defined categorisation of items based on their own tags.”
• Latent dirichlet allocation for tag recommendation (2009, background): “[19] propose exploiting co-ocurrence of users, resources, and tags for searching and ranking within tagging systems.”
• PageRank beyond the Web (2014, methodology): “FolkRank scores [Hotho et al., 2006] are designed to measure the importance of an object, tag, or user with respect to a small set of objects, tags, or users that define a topic.”
• Evaluating similarity measures for emergent semantics of social tagging (2009, methodology): “[23] convert a folksonomy into an undirected weighted network, used for computing a modified PageRank algorithm called FolkRank for ranking query results.”