Aleksandar Stupar

Learn More
We consider the problem of processing K-Nearest Neighbor (KNN) queries over large datasets where the index is jointly maintained by a set of machines in a computing cluster. The proposed RankReduce approach uses locality sensitive hashing (LSH) together with a MapReduce implementation, which by design is a perfect match as the hashing principle of LSH can(More)
In this work, a benchmark to evaluate the retrieval performance of soundtrack recommendation systems is proposed. Such systems aim at finding songs that are played as background music for a given set of images. The proposed benchmark is based on preference judgments, where relevance is considered a continuous ordinal variable and judgments are collected for(More)
We consider the task of automatically phrasing and computing top-k rankings over the information contained in common knowledge bases (KBs), such as YAGO or DBPedia. We assemble the thematic focus and ranking criteria of rankings by inspecting the present Subject, Predicate, Object (SPO) triples. Making use of numerical attributes contained in the KB we are(More)
Everything is relative. Cars are compared by gas per mile, websites by page rank, students based on GPA, scientists by number of publications, and celebrities by beauty or wealth. In this paper, we study the characteristics of such entity rankings based on a set of rankings obtained from a popular Web portal. The obtained insights are integrated in our(More)
Halls of Fame are fascinating constructs. They represent the elite of an often very large amount of entities—persons, companies, products, countries etc. Beyond their practical use as static rankings, changes to them are particularly interesting—for decision making processes, as input to common media or novel narrative science applications, or simply(More)
We demonstrate PICASSO, a novel approach to soundtrack recommendation. Given text, video, or image documents, PICASSO selects the best fitting music pieces, out of a given set of files, for instance, a user's personal mp3 collection. This task, commonly referred to as soundtrack suggestion, is non-trivial as it requires a lot of human attention and a good(More)
In this work, a benchmark to evaluate the retrieval performance of soundtrack recommendation systems is proposed. Such systems aim at finding songs that are played as background music for a given set of images. The proposed benchmark is based on preference judgments, where relevance is considered a continuous ordinal variable and judgments are collected for(More)
We describe the vision of a system that performs “educated guessing” to answer ad hoc information needs in case of missing or undisclosed information. The guessing procedure is based on discovered common patterns, obtained from structured and semi-structured data, guided by the specific information need.