IST LIVE

Content related to "content-recommendation"

(1) Recommender System for the Multi-Channel TV Production

This paper presents the concept of content recommendations for the production of multi-channel TV shows. Within the IST FP6 project "LIVE – Live Staging of Media Events" we are developing a production support system which will have a functionality of content recommendations and will support production of multi-channels programs. The paper outlines a concept of a recommender system for the multi-channel TV production and presents basic architecture and workflows within the system. The recommendation of the archive content for a given channel is personalized by taking into account the profile of the target audience.

(1) WP6 Status: User profiles, Feedback Analysis and content recommendations

Presentation of work carried out in WP6 and future work. Work on user modelling, feedback presentation, recommendations and issues and plans within wp6 are presented. Presented on PCC in Ljubljana on October 2007.

(1) Consumer Recommender and Feed-back demonstrator

The purpose of the iTV demonstrator is to show the possibilities that can be provided by ad-vanced iTV services. The first prototype described in this document is based on the back-ground architecture work developed at the Laboratory for digital signal, image and video processing , University of Ljubljana and on the JAME Author software, which was developed by the Fraunhofer-Institut für Medienkommunikation IMK (FhG). The prototype is used for demonstration of basic functionalities that are interesting for the coming iTV services such as video on demand, automatic recommendation of interesting TV programmes, acquisition of additional data for particular TV programme, etc.

(1) Personalised content search

An overview of different personalization approaches (content-based, collaborative.based, feedback gathering mechanisms, existing standards). How can this approaches be used in LIVE environment.

(1) The evaluation of a hybrid recommender system for recommendation of movies

In this paper we present our approach to generation of movie recommendations. The idea of our hybrid approach is to first separately generate predicted ratings for movies using the contentbased and collaborative recommender modules. Predicted ratings from both recommender engines are then combined into final classification by the hybrid recommender using weighted voting scheme. The basis for the calculations are Pearson’s correlation coefficient, True Bayesian prediction and M5Rules decision rules. The evaluation of the system performance was based on the EachMovie data corpus, for around 7000 users. Preliminary results show that this approach works really well, while there is still some room for improvement.

(1) Semantic-based Realization of Novel iTV Formats for the Broadcasting of Media Events (Full Paper)

Broadcasting of media events is a real-time action demanding reliable just in time decisions based on the current content of incoming video streams and the availability of background material. Novel iTV formats for broadcasting this type of event thus demand monitoring of multiple streams and background material. Due to the potentially large amount of streams and other available material, manual monitoring is likely to fail on the long term. We therefore developed an indexing pipeline based on semantic technologies that enables real-time analysis of broadcasted streams and reliable content recommendations of streams and background material based on formal machine understandable descriptions of content. Our approach enables real-time interpretation of broadcasted streams and thus establishes a bridge over the “Semantic Gap” in video analysis.

(1) Overview of LIVE Work Package 6, Personalisation and Feedback

Public overview of objectives and results of the work package 6, Personalisation and Feedback.

(1) An Analysis of New Feedback Methods for Parallel Multi-Stream Productions

In this work we explore the usage of novel real-time feedback tools in the context of a new iTV production process, developed within the scope of the research project LIVE. The exploration is based on analysis of log-files, observations and interviews captured during a field trial at the Austrian broadcaster ORF. We work out characteristic usage patterns, as well as the broadcaster's and end user's acceptance. In doing so, we give an insight into the possibilities of feedback tools with regard to parallel multi-stream productions.

(0.7) Personal content recommender based on a hierarchical user model for the selection of TV programmes

In this paper we present our approach to user modeling for a personalized selection of multimedia content tested on a corpus of TV programmes. The idea of this approach is to classify content (TV programmes) based on the calculation of similarities between the description of content and the user model for each descrip- tion attribute. Calculated similarities are then combined into a classi¯cation decision using the Support Vector Machines. The basis for the calculation of similarities is a hierarchical structure of the user model, overlaid upon a taxonomy of TV pro- gramme genres. Preliminary results show that it works well with a varying quality of content descriptions including incomplete genre classi¯cation and arbitrary number of description attributes. The evaluation of the system performance was based on content described using the TV-Anytime standard, but the approach can be adapted for search of other types of content with multi-attribute descriptions.

(0.7) D9.6 Report on requirement monitoring

As the final document of WP9 this report lists various revised issues that have been described in separate deliverables in former times. Chapter 3 depicts the relevant actors and roles within the LIVE system. The final usage scenarios that could be observed during the preparation, production and post-processing of the Olympic Trial are described in chapter 4. They were the precondition for the revision of the final requirements presented in chapter 5. The description of the final requirements is oriented towards the presentation of the revised requirements in [LIVE D9.4] and follows the same outline. They again served as a foundation for the revision of the final system architecture of the LIVE production support system. Chapter 6 picks up the already defined workflows from [LIVE D9.5] and adapts and amends them according to the final requirements from chapter 5. This is accomplished by providing a componentbased view on the one hand and more detailed inspections on components which have been subject to major changes on the other. The first part intends to give an overview on the performed adaptations of the architecture for each component. The latter part is meant to serve as a guideline for the developers of the various components and tools on which interfaces will be needed between which components and which data has to be exchanged.

(0.7) D8.6 Report on the integration of the second system prototype

This document reports on the process of integration, testing and operation of the second prototype version of the LIVE System. It presents the complexity of the system including the implemented internal and external interfaces. In order to verify the development of the required and agreed function set and to ensure a certain level of stability, scalability and performance, extensive system testing was performed. The chosen approach, its realisation and conclusion comprise a major part of the report. In contrast to the first system prototype this version was foreseen to come to operation during the Summer Olympics 2008. A production team at the ORF in Vienna should be supported in their task to produce a multichannel interactive live programme. The operation of the system during the Olympic field trial is summarised from an integration point of view.