Stefan Dobravec

Share it! - content transfer in home-to-home networks

This paper presents a possible solution for the introduction of a complementary distribution chain into the broadcast environment. In the described scenario, personal video recorders with local storage are networked into an ad-hoc home-to-home network based on an emerging peer-to-peer protocol that enables broadcast as well as home-produced content sharing. The paper presents the current system architecture as conceived within a consortium working on the Share it! project.

VICTORY: a 3D search engine over P2P and wireless P2P networks

Our information society is being overflowed with digital data of various kinds. In order for the users to find relevant information, novel digital repositories should provide search mechanisms that are both content and context based, facilitating the formulation of queries and enabling search by example using 3D/2D objects, sketches, etc. At the same time the repositories should scale to the growing number of users. This paper presents the main achievements of the IST project VICTORY that aims at developing an innovative distributed digital repository introducing novel search, retrieval, and visualization capabilities to a desktop and mobile P2P networks.

Peer to peer search engine and collaboration platform based on JXTA protocol

In this paper we are going to present the architecture of peer-to-peer collaborative framework. The peer-to-peer technologies used in this research are based on Java implementation of JXTA protocol. The communication services that are implemented on presented peer-to-peer network do not depend on centralized servers and are therefore immune to server infrastructure faults. Moreover an advantage of the presented approach is the ability to seamlessly establish communication through firewalls and proxy servers (e.g. NAT server).

Increasing Top-20 Diversity Through Recommendation Post-processing

This paper presents two different methods for diversifying recommendations that were developed as part of the ESWC2014 chal- lenge. Both methods focus on post-processing recommendations provided by the baseline recommender system and have increased the ILD at the cost of final precision (measured with F@20). The authors feel that this method has potential yet requires further development and testing.

TIRAMISU: A novel approach to content representation and key management for seamless super-distribution of protected media

Abstract Media delivery over heterogeneous networks requires both flexible representation and robust protection of content. This paper provides details on the framework for audiovisual content creation, delivery, consumption and protection as conceived within the IST project The Innovative Rights and Access Management Interplatform SolUtion. The proposed framework is based on the emerging MPEG-21 standard for multimedia content delivery and consumption and at the same time it complements it in several aspects, most notably by fully specifying a digital rights management (DRM) scheme. Central to the described framework is a novel key management system, relying on smartcards, which addresses many issues that previously blocked wider adoption of DRM: obtrusiveness of the DRM technology perceived by the end-user, flexibility in licence formulation and adequate level of trust as requested by content owners.

Two-way video transmission trial over SHDSL

This paper discusses two-way video transmission over SHDSL. We have tested the performance of the SHDSL modems by establishing real-time video sessions at data rates up to 6408 kbps. The range of applications that could benefit from the deployment of SHDSL technology is very broad including video on demand applications, video conferencing, remote video education applications, etc. SHDSL modems can achieve symmetrical data transfers at bit rates as high as 4608 kbps using copper-wire infrastructure.

Using latent features to measure the diversity of recommendation lists

One of the well-known issues with content recommender system is that they tend to become over-specialized, which often has a negative influence on user experience. This can be solved by diversification of the recommendation list, a process that implements a tradeoff between accuracy and diversity of recommended items. Normally, item metadata is used in the diversity measure. In certain cases however, the item metadata may not be available thus a different approach to measure diversity is required. The aim of this preliminary study is to determine whether latent features can be used to measure the diversity of recommended items. In order to resolve this we generated recommendation lists for 43 different users using the LDOS-CoMoDa dataset. We then evaluated the diversity of these lists using the standard intra-list diversity measure. In addition we calculated the diversity of each list by comparing the latent features (calculated using the matrix factorization approach) of each item on the list. The comparison of both value sets showed that they show similar characteristics which implies that latent feature space offers an alternative method of evaluating item diversity when no metadata is present.

The MUFFINS online store - buy it in the MPEG-21 way

Multimedia technology is being developed in light-speed velocity. The number of different players in the multimedia value and delivery chain is constantly increasing. New multimedia services and applications are emerging daily and furthermore, they are evolving constantly. However, the speed of the development, mass of information, numerous information providers and consumers, pose a great threat to the development of multimedia technology. This can be abridged by an adherence to the open standards, such as MPEG-21 multimedia framework. MPEG-21 is trying to set up an abstract model, which could unambiguously describe the multimedia content, allow seamless access and delivery of the content including the intellectual property management and protection mechanisms. Even more, the abstract model being developed is designed to allow future technologies to be implemented within. MPEG-21 promises to solve several issues within the domain of multimedia content distribution. Nevertheless this remains to be proven through implementation and wide acceptance. Consequently, the MUFFINS (multimedia framework for interoperability in secure environments) project has started to closely follow and monitor the development of the MPEG-21 standard, also implemented in an MPEG-21 real-life scenario. The goal of this study is to prove the MPEG-21 concept and to show its impact on the multimedia technology development as well as to monitor and contribute to the development of MPEG-21 standard. In this article, the basic concepts of the MPEG-21 standard and the MUFFINS project are presented.

Predicting sports results using latent features: A case study

Predicting sports results is normally a challenging task, even more in case of a sport that shows a highly stochastic nature. In football, for example, numerous features are tracked and combined with expert knowledge, yielding various predicting algorithms. Our work however, is based on a case where there is no expert knowledge available and the only data comes from previous match results. We built a goal score prediction model that uses latent features obtained from matrix factorization process. We also added a Naive Bayes Classifier to be able to predict outcome of the match. The algorithm has been tested on results of the FIFA World Cup 2014. We also built a match result predictor based on the betting quotas. As these are derived from a complex algorithms that encompass also the expert knowledge, our algorithm can be used to estimate accuracy of an expert knowledge-based system. This case study shows that there is no significant difference between the two algorithms that we tested and that the latent features may provide a valid substitute for real features, when the later ones are not available.