In collaborative tagging systems, it is generally assumed that users assign tags to facilitate retrieval of content at a later time. There is, however, little behavioral evidence that tags actually serve this purpose. Using a large-scale dataset from the social music website Last.fm, we explore how patterns of music tagging and subsequent listening interact to determine if there exist measurable signals of tags functioning as retrieval aids. Specifically, we describe our methods for testing if the assignment of a tag tends to lead to an increase in listening behavior. Results suggest that tagging, on average, leads to only very small increases in listening rates, and overall the data do

Social tagging systems have established themselves as an important part in today's web and have attracted the interest from our research community in a variety of investigations. The overall vision of our community is that simply through interactions with the system, i.e., through tagging and sharing of resources, users would contribute to building useful semantic structures as well as resource indexes using uncontrolled vocabulary not only due to the easy-to-use mechanics. Henceforth, a variety of assumptions about social tagging systems have emerged, yet testing them has been difficult due to the absence of suitable data. In this work we thoroughly investigate three available assumptions - e.g., is a tagging system really social? - by examining live log data gathered from the real-world public social tagging system BibSonomy. Our empirical results indicate that while some of these assumptions hold to a certain extent, other assumptions need to be reflected and viewed in a very critical light. Our observations have implications for the design of future search and other algorithms to better reflect the actual user behavior.

The information contained in social tagging systems is often modelled as a graph of connections between users, items and tags. Recommendation algorithms such as FolkRank, have the potential to leverage complex relationships in the data, corresponding to multiple hops in the graph. We present an in-depth analysis and evaluation of graph models for social tagging data and propose novel adaptations and extensions of FolkRank to improve tag recommendations. We highlight implicit assumptions made by the widely used folksonomy model, and propose an alternative and more accurate graph-representation of the data. Our extensions of FolkRank address the new item problem by incorporating content data into the algorithm, and significantly improve prediction results on unpruned datasets. Our adaptations address issues in the iterative weight spreading calculation that potentially hinder FolkRank's ability to leverage the deep graph as an information source. Moreover, we evaluate the benefit of considering each deeper level of the graph, and present important insights regarding the characteristics of social tagging data in general. Our results suggest that the base assumption made by conventional weight propagation methods, that closeness in the graph always implies a positive relationship, does not hold for the social tagging domain.

With the rising popularity of smart mobile devices, sensor data-based applications have become more and more popular. Their users record data during their daily routine or specifically for certain events. The application WideNoise Plus allows users to record sound samples and to annotate them with perceptions and tags. The app is being used to document and map the soundscape all over the world. The procedure of recording, including the assignment of tags, has to be as easy-to-use as possible. We therefore discuss the application of tag recommender algorithms in this particular scenario. We show, that this task is fundamentally different from the well-known tag recommendation problem in folksonomies as users do no longer tag fix resources but rather sensory data and impressions. The scenario requires efficient recommender algorithms that are able to run on the mobile device, since Internet connectivity cannot be assumed to be available. Therefore, we evaluate the performance of several tag recommendation algorithms and discuss their applicability in the mobile sensing use-case.

FReSET is a new recommender systems evaluation framework aiming to support research on folksonomy-based recommender systems. It provides interfaces for the implementation of folksonomy-based recommender systems and supports the consistent and reproducible offline evaluations on historical data. Unlike other recommender systems framework projects, the emphasis here is on providing a flexible framework allowing users to implement their own folksonomy-based recommender algorithms and pre-processing filtering methods rather than just providing a collection of collaborative filtering implementations. FReSET includes a graphical interface for result visualization and different cross-validation implementations to complement the basic functionality.

Social resource sharing systems are central elements of the Web 2.0 and use the same kind of lightweight knowledge representation, called folksonomy. Their large user communities and ever-growing networks of user-generated content have made them an attractive object of investigation for researchers from different disciplines like Social Network Analysis, Data Mining, Information Retrieval or Knowledge Discovery. In this paper, we summarize and extend our work on different aspects of this branch of Web 2.0 research, demonstrated and evaluated within our own social bookmark and publication sharing system BibSonomy, which is currently among the three most popular systems of its kind. We structure this presentation along the different interaction phases of a user with our system, coupling the relevant research questions of each phase with the corresponding implementation issues. This approach reveals in a systematic fashion important aspects and results of the broad bandwidth of folksonomy research like capturing of emergent semantics, spam detection, ranking algorithms, analogies to search engine log data, personalized tag recommendations and information extraction techniques. We conclude that when integrating a real-life application like BibSonomy into research, certain constraints have to be considered; but in general, the tight interplay between our scientific work and the running system has made BibSonomy a valuable platform for demonstrating and evaluating Web 2.0 research.

This paper describes our approach to the ECML/PKDD Discovery Challenge 2009. Our approach is a pure statistical model taking no content information into account. It tries to find latent interactions between users, items and tags by factorizing the observed tagging data. The factorization model is learned by the Bayesian Personal Ranking method (BPR) which is inspired by a Bayesian analysis of personalized ranking with missing data. To prevent overfitting, we ensemble the models over several iterations and hyperparameters. Finally, we enhance the top-n lists by estimating how many tags to recommend.

Tag recommendation is the task of predicting a personalized list of tags for a user given an item. This is important for many websites with tagging capabilities like last.fm or delicious. In this paper, we propose a method for tag recommendation based on tensor factorization (TF). In contrast to other TF methods like higher order singular value decomposition (HOSVD), our method RTF ('ranking with tensor factorization') directly optimizes the factorization model for the best personalized ranking. RTF handles missing values and learns from pairwise ranking constraints. Our optimization criterion for TF is motivated by a detailed analysis of the problem and of interpretation schemes for the observed data in tagging systems. In all, RTF directly optimizes for the actual problem using a correct interpretation of the data. We provide a gradient descent algorithm to solve our optimization problem. We also provide an improved learning and prediction method with runtime complexity analysis for RTF. The prediction runtime of RTF is independent of the number of observations and only depends on the factorization dimensions. Besides the theoretical analysis, we empirically show that our method outperforms other state-of-the-art tag recommendation methods like FolkRank, PageRank and HOSVD both in quality and prediction runtime.

In this paper we consider the problem of item recommendation in collaborative tagging communities, so called folksonomies, where users annotate interesting items with tags. Rather than following a collaborative filtering or annotation-based approach to recommendation, we extend the probabilistic latent semantic analysis (PLSA) approach and present a unified recommendation model which evolves from item user and item tag co-occurrences in parallel. The inclusion of tags reduces known collaborative filtering problems related to overfitting and allows for higher quality recommendations. Experimental results on a large snapshot of the <i>delicious</i> bookmarking service show the scalability of our approach and an improved recommendation quality compared to two-mode collaborative or annotation based methods.

Folksonomy systems have shown to contribute to the quality of Web search ranking strategies. In this paper, we analyze and compare different graph-based ranking algorithms, namely FolkRank, SocialPageRank, and SocialSimRank. We enhance these algorithms by exploiting the context of tag assignmets, and evaluate the results on the GroupMe! dataset. In GroupMe!, users can organize and maintain arbitrary Web resources in self-defined groups. When users annotate resources in GroupMe!, this can be interpreted in context of a certain group. The grouping activity delivers valuable semantic information about resources and their context. We show how to use this information to improve the detection of relevant search results, and compare different strategies for ranking result lists in folksonomy systems.

Collaborative tagging systems allow users to assign keywords - so called "tags" - to resources. Tags are used for navigation, finding resources and serendipitous browsing and thus provide an immediate benefit for users. These systems usually include tag recommendation mechanisms easing the process of finding good tags for a resource, but also consolidating the tag vocabulary across users. In practice, however, only very basic recommendation strategies are applied. In this paper we evaluate and compare several recommendation algorithms on large-scale real life datasets: an adaptation of user-based collaborative filtering, a graph-based recommender built on top of the FolkRank algorithm, and simple methods based on counting tag occurences. We show that both FolkRank and Collaborative Filtering provide better results than non-personalized baseline methods. Moreover, since methods based on counting tag occurrences are computationally cheap, and thus usually preferable for real time scenarios, we discuss simple approaches for improving the performance of such methods. We show, how a simple recommender based on counting tags from users and resources can perform almost as good as the best recommender.

Social bookmarking systems allow users to store links to internet resources on a web page. As social bookmarking systems are growing in popularity, search algorithms have been developed that transfer the idea of link-based rankings in the Web to a social bookmarking system's data structure. These rankings differ from traditional search engine rankings in that they incorporate the rating of users.</p> <p>In this study, we compare search in social bookmarking systems with traditionalWeb search. In the first part, we compare the user activity and behaviour in both kinds of systems, as well as the overlap of the underlying sets of URLs. In the second part,we compare graph-based and vector space rankings for social bookmarking systems with commercial search engine rankings.</p> <p>Our experiments are performed on data of the social bookmarking system Del.icio.us and on rankings and log data from Google, MSN, and AOL. We will show that part of the difference between the systems is due to different behaviour (e. g., the concatenation of multi-word lexems to single terms in Del.icio.us), and that real-world events may trigger similar behaviour in both kinds of systems. We will also show that a graph-based ranking approach on folksonomies yields results that are closer to the rankings of the commercial search engines than vector space retrieval, and that the correlation is high in particular for the domains that are well covered by the social bookmarking system.

Social bookmarking systems constitute an established part of the Web 2.0. In such systems users describe bookmarks by keywords called tags. The structure behind these social systems, called folksonomies, can be viewed as a tripartite hypergraph of user, tag and resource nodes. This underlying network shows specific structural properties that explain its growth and the possibility of serendipitous exploration. Today’s search engines represent the gateway to retrieve information from the World Wide Web. Short queries typically consisting of two to three words describe a user’s information need. In response to the displayed results of the search engine, users click on the links of the result page as they expect the answer to be of relevance. This clickdata can be represented as a folksonomy in which queries are descriptions of clicked URLs. The resulting network structure, which we will term logsonomy is very similar to the one of folksonomies. In order to find out about its properties, we analyze the topological characteristics of the tripartite hypergraph of queries, users and bookmarks on a large snapshot of del.icio.us and on query logs of two large search engines. All of the three datasets show small world properties. The tagging behavior of users, which is explained by preferential attachment of the tags in social bookmark systems, is reflected in the distribution of single query words in search engines. We can conclude that the clicking behaviour of search engine users based on the displayed search results and the tagging behaviour of social bookmarking users is driven by similar dynamics.

Folksonomies in Wissensrepräsentation und Information Retrieval. Die populären Web 2.0-Dienste werden von Prosumern -- Produzenten und gleichsam Konsumenten -- nicht nur dazu genutzt, Inhalte zu produzieren, sondern auch, um sie inhaltlich zu erschließen. Folksonomies erlauben es dem Nutzer, Dokumente mit eigenen Schlagworten, sog. Tags, zu beschreiben, ohne dabei auf gewisse Regeln oder Vorgaben achten zu müssen. Neben einigen Vorteilen zeigen Folksonomies aber auch zahlreiche Schwächen (u. a. einen Mangel an Präzision). Um diesen Nachteilen größtenteils entgegenzuwirken, schlagen wir eine Interpretation der Tags als natürlichsprachige Wörter vor. Dadurch ist es uns möglich, Methoden des Natural Language Processing (NLP) auf die Tags anzuwenden und so linguistische Probleme der Tags zu beseitigen. Darüber hinaus diskutieren wir Ansätze und weitere Vorschläge (Tagverteilungen, Kollaboration und akteurspezifische Aspekte) hinsichtlich eines Relevance Rankings von getaggten Dokumenten. Neben Vorschlägen auf ähnliche Dokumente (glqqmore like this!grqq) erlauben Folksonomies auch Hinweise auf verwandte Nutzer und damit auf Communities (glqqmore like me!grqq). Folksonomies in Knowledge Representation and Information Retrieval In Web 2.0 services grqqprosumers” -- producers and consumers -- collaborate not only for the purpose of creating content, but to index these pieces of information as well. Folksonomies permit actors to describe documents with subject headings, grqqtagsgrqq, without regarding any rules. Apart from a lot of benefits folksonomies have many shortcomings (e.g., lack of precision). In order to solve some of the problems we propose interpreting tags as natural language terms. Accordingly, we can introduce methods of NLP to solve the tags’ linguistic problems. Additionally, we present criteria for tagged documents to create a ranking by relevance (tag distribution, collaboration and actor-based aspects). Besides recommending similar documents (glqqmore like this!grqq) folksonomies can be used for the recommendation of similar users and communities (glqqmore like me!grqq).

Social bookmark tools are rapidly emerging on the Web. In such systems users are setting up lightweight conceptual structures called folksonomies. The reason for their immediate success is the fact that no specific skills are needed for participating. In this paper we specify a formal model for folksonomies and briefly describe our own system BibSonomy, which allows for sharing both bookmarks and publication references in a kind of personal library.