Tom Crecelius

Efficient top-k querying over social-tagging networks

Online communities have become popular for publishing and searching content, as well as for finding and connecting to other users. User-generated content includes, for example, personal blogs, bookmarks, and digital photos. These items can be annotated and rated by different users, and these social tags and derived user-specific scores can be leveraged for searching relevant content and discovering subjectively interesting items. Moreover, the relationships among users can also be taken into consideration for ranking search results, the intuition being that you trust the recommendations of your close friends more than those of your casual acquaintances. Queries for tag or keyword combinations that compute and rank the top-k results thus face a large variety of options that complicate the query processing and pose efficiency challenges. This paper addresses these issues by developing an incremental top-k algorithm with two-dimensional expansions: social expansion considers the strength of relations among users, and semantic expansion considers the relatedness of different tags. It presents a new algorithm, based on principles of threshold algorithms, by folding friends and related tags into the search space in an incremental on-demand manner. The excellent performance of the method is demonstrated by an experimental evaluation on three real-world datasets, crawled from deli.cio.us, Flickr, and LibraryThing.

Exploiting social relations for query expansion and result ranking

Online communities have recently become a popular tool for publishing and searching content, as well as for finding and connecting to other users that share common interests. The content is typically user-generated and includes, for example, personal blogs, bookmarks, and digital photos. A particularly intriguing type of content is user-generated annotations (tags) for content items, as these concise string descriptions allow for reasonings about the interests of the user who created the content, but also about the user who generated the annotations. This paper presents a framework to cast the different entities of such networks into a unified graph model representing the mutual relationships of users, content, and tags. It derives scoring functions for each of the entities and relations. We have performed an experimental evaluation on two real-world datasets (crawled from deli.cio.us and Flickr) where manual user assessments of the query result quality show that our unified graph framework delivers high-quality results on social networks.

Pay-as-you-go maintenance of precomputed nearest neighbors in large graphs

An important building block of many graph applications such as searching in social networks, keyword search in graphs, and retrieval of linked documents is retrieving the transitive neighbors of a node in ascending order of their distances. Since large graphs cannot be kept in memory and graph traversals at query time would be prohibitively expensive, the list of neighbors for each node is usually precomputed and stored in a compact form. While the problem of precomputing all-pairs shortest distances has been well studied for decades, efficiently maintaining this information when the graph changes is not as well understood. This paper presents an algorithm for maintaining nearest neighbor lists in weighted graphs under node insertions and decreasing edge weights. It considers the important case where queries are a lot more frequent than updates, and presents two approaches for transparently performing necessary index updates while executing queries. Extensive experiments with large graphs, including a subset of Twitters user graph, demonstrate that the overhead for this maintenance is small.

Social recommendations at work

Online communities have become popular for publishing and searching content, and also for connecting to other users. User-generated content includes, for example, personal blogs, bookmarks, and digital photos. Items can be annotated and rated by different users, and users can connect to others that are usually friends and/or share common interests. We demonstrate a social recommendation system that takes advantages of users connections and tagging behavior to compute recommendations of items in such communities. The advantages can be verified via comparison to a standard IR technique.