An Exploratory Study on the Explicitness of User Intentions in Digital Photo Retrieval

Search queries are typically interpreted as specification of information need of a user. Typically the search query is either interpreted as is or based on the context of a user, being for instance a user profile, his/her previously undertaken searches or any other background information. The actual intent of the user – the goal s/he wants to achieve with information retrieval – is an important part of a user’s context. In this paper we present the results of an exploratory study on the interplay between the goals of users and their search behavior in multimedia retrieval.

Using Visual Features to Improve Tag Suggestions in Image Sharing Sites

Social media sharing sites such as Flickr or YouTube have become immensely popular. Besides sharing actual content, users also share annotations describing or classifying the contents they publish. Although tagging is easy, annotation still is a laborious task that can be made easier by suggesting meaningful additional tags to the user automatically. In this position paper we propose a system architecture and process for supporting annotation by tag suggestion to increase the quality and quantity of social annotations. The goal is not to tag previously untagged images in a completely automatic way, but instead to extend the amount and completeness of annotations by supporting the user in the process of adding further tags.

Getting to “Know” People on the Web 2.0

Web 2.0 platforms such as media sharing and social network sites (SNS) concern people in everyday life to a great extent. People are enabled to reach out to various media and up to now, it is nearly impossible to use digital identities ex ante or to recreate users’ identities ex post across different platforms. In this paper, we explore important methodologies in Web 2.0 such as cross-media analysis and social pattern based analysis based on a survey in this area, aiming at cross-platform information diffusion across social network sites. Open issues are discussed to explore the challenges and solutions in this new research area.

Virtual Campfire – A Mobile Social Software for Cross-Media Communities

Multimedia creation, annotation and sharing are challenging tasks especially of interdisciplinary, intercultural and intergenerational communities. We present the mobile social software Virtual Campfire to provide cross-media and cross-community support for de- and recontextualization of multimedia content, employing Web Services, the MPEG-7 standard and Web 2.0 technologies etc. Virtual Campfire can enable communities to set up and maintain multimedia community information systems quickly and easily.

MPEG-7 for Video Quality Description and Summarisation

Manual quality control of audiovisual content in the different steps of the media production, delivery and archiving process causes significant costs. Semi-automatic quality control requires automatisation of quality analysis, quality metadata interoperability and efficient visualisation tools. In this paper we propose the use of MPEG-7 for standard compliant description of media quality metadata and a quality summary visualisation tool which facilitates efficient exploration of visually impaired content by the user.

Personalizing the Web Content on User Perceptual Preferences

This paper introduces a new model of personalized usage of the internet that is based on technologies of user representation, artificial intelligence and semantic augmentation of the content. By taking advantage of internet’s unprecedented dynamics, compared to traditional media, this user representation model incorporates cognitive, mainly, psychology theories, combined with parameters that constitute more traditional approaches in user profiling (such as demographics, expertise, etc). The purpose of this research is to alleviate difficulties that massive approaches impose on areas such as education and information processing, by integrating intelligent adaptive characteristics into web applications; this can lead to a highly adapted to each user’s needs content and more effective, in our case, learning.

Enhancing Music Learning Experience through a Dedicated Web 2.0 Oriented Service

The Web 2.0 philosophy has brought new ways of using the web as a content repository and a sharing platform. Non-computer skilled people can now publish their own text, images, videos and/or sounds and take part in communities created around topics they like. Our idea is thus to use this new communication mechanism to assist skill learning. Skill is indeed a difficult knowledge to learn on a text form as it is hard to describe movements, gestures or procedures in this way: sometime, a picture or a video is better than a  thousand of words.
As a popular field of experience, we have focused our attention on music learning, and more particularly on guitar pieces learning. Music is very representative of skill learning, it is both a physical and an intellectual activity. The “Gloss2U” service we describe in this article takes into account the specificities of this learning process, especially as events are time-related.

To achieve this, we rely on new equipment conditions that are nowadays gathered. Userfriendly multimedia tools opened new horizons and broadband networks (ADSL, cable, …) are becoming more and more common. Almost everybody is a potential content producer with just a webcam or a cell phone. Our system is therefore audio and video based and users interact with each others by submitting contributions, called glosses, in the context of the piece they are learning. A dialogue starts between student users (learners) and experimented ones (professors), to discuss encountered problems and the way to solve them. The other strength of this system is that it can act as a knowledge repository for forthcoming practitioners.

Knowledge Management from an Ancient Chinese Point of View, or the Knowledge Flow through an Organisation

The ancient Chinese kind of thinking, as best known from TCM (Traditional Chinese Medicine) is very different from the modern western approach of scientific medicine. But there is powerful knowledge in both systems and the knowledge is complementary. It is obvious that our methods and  techniques of knowledge management do not support the foreign kind of thinking. Though many of our trends have striking similarity to this ancient  Chinese behaviour. Holistic approaches, synthesis versus analysis, networking, social web, empathy, mirror neurons, memetics, graphical versus textual communication. Many of the things we cannot understand are complex systems and for these systems our analytic behaviour in thinking does not seem to work. The straight rule of cause and effect does not work for complex systems, so our western philosophy based on causality fails. Our attempt in managing knowledge using analytical models based on causality will probably fail too. To break up information into sentences of words and looking for grammar (taxonomy), syntax and relations (ontology) cannot work. Because it is the wrong way of asking questions compared to the ancient Chinese philosophy. So what could be a useful approach in managing complex systems like knowledge in organisations from the ancient Chinese point of view?