Information & Communication Technologies

 
Path:
 Home > Programme > Challenge4 > Content Knowledge > Projects > WeKnowIt
Site navigation:

Technologies for Information Management

Project

WEKNOWIT logo

WeKnowIt – Emerging, Collective Intelligence for personal, organisational and social use

Novel techniques for generating different layers of intelligence

Due to advances in communications, mobile devices and Web technologies, it is nowadays easy for users and organisations to generate and share content, individually or within communities. Yet, current applications do not fully support intelligent processing and management of such information.

The main objective of WeKnowIt is to develop novel techniques for exploiting multiple layers of intelligence from user-contributed content, which together constitute Collective Intelligence, a form of intelligence that emerges from the collaboration and competition among many individuals, and that seemingly has a mind of its own.

To this end, input from various sources is analysed and combined: from digital content items and contextual information (Media Intelligence), massive user feedback (Mass Intelligence), and users social interaction (Social Intelligence) so as to benefit end-users (Personal Intelligence) and organisations (Organisational Intelligence).

WeKnowIt will demonstrate the wide applicability of its achievements through the elaboration on two distinct case studies: an Emergency Response and a Consumers Social Group case study. The two case studies pertain to different topics, target at a wide range of intended users and involve heterogeneous business models so that WeKnowIt achieves the highest possible social impact.

The driving vision behind WeKnowIt is the opportunity to analyse user contributed content by integrating research and development in personalisation, content processing, user feedback, social analysis and knowledge management to automatically generate Collective Intelligence and make it accessible to end users and organisations. Currently Web 2.0 methodologies miss an 'Intelligence' layer over existing manually generated information that would enable the exploitation of the knowledge hidden in the user contributed content. Although current Web 2.0 applications allow and are based on annotations by the users or 'tags', these are not sufficient because user generated tags lack clear semantics, therefore these social annotations are not of much use for web agents and applications. WeKnowIt will not just try to automate the content annotation processing but will also exploit and analyze all available information related to user submitted content by researching novel methods for content processing and analysis, analyse mass user feedback and social interactions. The WeKnowIt approach is built around different Intelligence Layers which address various aspects of user contributed and consumed content. Users submitting content will benefit from the advances of WeKnowIt for bringing Web 2.0 and user contributed content applications to users of mobile devices.

The Personal Intelligence layer will enable users to efficiently submit their content but it also involves modelling and extraction of user preferences, enabling personalised access to the content and knowledge available from the WeKnowIt applications. Once content is submitted, a series of processing and analysis procedures take place in order to exploit all available sources of information.

The Media Intelligence layer applies semantic analysis to the content items, taking into account the content itself (text, images, video, speech), existing tags, personal, social and contextual information (e.g. position, time). This way, a number of semantic, consistent, machine-processable tags are added to the already user defined tags, while the existing tags through reasoning services are checked for consistency and are corrected.

Mass Intelligence combines the information from mass user feedback in order to extract patterns and trends that cannot be extracted by single content items. This adds an additional layer of semantic description to the content or can describe overall situations.

In parallel, Social Intelligence analyses the usage and communication interaction patterns taking into account the needs and capabilities of communities and provides useful output both for the creation of the Media and Mass as well as for the WeKnowIt applications. For example, the identification of 'authorities' among users can be of benefit for Media and Mass Intelligence creation, since the analysis can place more emphasis on the content of such users. At the same time, WeKnowIt content pushed to 'hub' users ensures faster distribution within communities.

After processing takes places in a continuous interaction way during the creation of Media, Mass and Social Intelligence, the generated Collective Intelligence comprising of all Intelligence Layers is made available to the users through the Personal Intelligence to end users and the Organisational Intelligence to users of organisations. Organisational Intelligence allows support of decision making through workflows exploiting the knowledge generated by WeKnowIt and taking into account existing procedures within the organisation.

The aggregated benefits of Collective Intelligence acquired through the various Intelligence Layers are, in particular, realised by both the end users and the organisations when uploading, searching for, browsing and consuming the content. Realisation of this value comes from an integrated project, where technology development for all parts can be accelerated via the collaboration between the knowledge and multimedia technology domains. Each research activity in WeKnowIt does not only contribute to the generation of Collective Intelligence resulting from each innovation separately, but the integration of these varied activities results in an overall project innovation which exceeds the aggregate of the individual innovation parts.



WeKnowIt Collective Intelligence



Technical details

WeKnowIt consists of five layers of Intelligence that are: Personal Intelligence, Media Intelligence, Mass Intelligence, Social Intelligence and Organisational Intelligence (Figure 1). A short description of the five layers is presented below.


Layers of Intelligence in WeKnowIt

Figure 1 : Layers of Intelligence in WeKnowIt


Personal Intelligence
Methodologies and technologies will be designed, developed and tested for Personal Intelligence management where citizens or users are enabled to interact with the WeKnowIt system in terms of both uploading information and accessing available information. WeKnowIt will enable efficient interaction using devices such as personal computers (e.g. via internet), but also devices with limited capabilities in terms of interfaces (e.g. mobile phones, PDAs, etc.). A main requirement is the creation of an interaction modality able to anticipate usersâ?? needs in order to enable effective use for user interaction, especially in emergency situations where time is limited and attention can be easily diverted. Strategies and modalities of interaction which will enable anticipation of user actions (hence reducing the interaction complexity) as well as a set of strategies to maintain and guarantee privacy. Usability will be empowered by content based analysis of input (images, text, videos), as well as on social, community and organisational analysis which will lead to more autonomous content actions, thereby reducing the interaction burden on the user.

Media Intelligence
Most of related knowledge and information originates from raw content, be it in the form of e.g. text, images, video, or speech. Human annotation or tagging used in social networks is a way to represent or handle the underlying knowledge, yet despite the human intervention, content remains highly unstructured and it is quite difficult to extract semantics and correlate to other sources of information.
Methods for single and cross-media analysis will use prior knowledge, either implicit, in the form of supervised learning from training data, or explicit, in the form of knowledge driven approaches to maximise extraction and overcome limitations of the current technology.
Novel methods of fusing information from different sources/modalities, contextual information (e.g. time, location, acquisition metadata), personal context (profile, preferences, etc.) and social context (tagging, ratings, group profiles, relevant content collections etc.).

Mass Intelligence
Mass Intelligence is recognition and understanding of facts and trends by exploitation of massive user contributions. Mass Intelligence is always useful when the aggregation of data, metadata and behaviour from and of a large mass of users gives new insight that would not be possible by investigating the contributions at the individual level only. WeKnowIt will investigate four aspects of Mass Intelligence:
  1. Mass question answering: How does question answering by masses of users give improved insights?
  2. Mass interaction feedback: How does explicit and implicit feedback by users during interaction may be gathered and analyzed to improve overall importance ratings and rankings of media and metadata?
  3. Mass classification and clustering: How may classifications of individuals be aggregated into meaningful overall classifiers and clusters of media and facts?
  4. Mass evolution analysis: How may a change in the behaviour and usage of data affect the understanding of media and facts?

Social Intelligence
WeKnowIt will also target the analysis, recognition and understanding of needs and capabilities of communities and communication interaction patterns. It will exploit existing work in Social Network Analysis and the dynamics of social systems improving upon state-of-the-art in terms of visualisation, navigation, as well as community analysis and management techniques. More specifically, WeKnowIt will aim:
  1. To improve community knowledge by the cross-usage of Personal, Mass and Organisational Intelligence as well as to provide community knowledge to fertilise Personal, Mass and Organisational Intelligence.
  2. To develop an intuitive, easy to use, flexible and powerful community administration platform that supports self organisation of communities and a wide range of privacy protection policies.
  3. To develop scalable community analysis tools, which are context sensitive and adhere to privacy protection policies.
  4. To communicate efficiently community structure and knowledge in line with existing privacy policies.
  5. To validate and demonstrate the feasibility of community services in the case studies under operating conditions.

Organisational Intelligence
Finally, WeKnowIt will bring the innovation of Web 2.0 technologies to the field of Organisational Intelligence where processes and workflows are set up in order to bring the right piece of knowledge at the right time to the right person in the organisation in order to support decision making. In WeKnowIt, however, this knowledge is not necessarily produced by the individual knowledge worker, but rather by the interaction with the layers of Personal, Media, Mass and Social Intelligence. The objective is to research new ways of setting up such a layer of Organisational Intelligence as a mixture of organisational structuring (social hierarchies, workflows, groups of interest, person roles) and self-organizing intelligence.

More details

Events in connection with WeKnowIt 

 


Project coordinators:

Dr Yiannis Kompatsiaris, Informtics and Telematics Institute, Multimedia Knowledge Lab, Greece
Email to Dr Kompatsiaris

Dr Yannis Avrithis, Video and Multimedia Systems Laboratory, School of Electrical and Computer Engineering, National Technical University of Athens, Greece
Email to Dr Yannis Avrithis



Participants
CERTH - ITI, Greece (coordinator)
Vysoke Uceni Technicke V Brne, Czech Republic
Lycos Europe GmbH, Germany
Universitaet Koblenz-Landau, Germany
Universitaet Karlsruhe (TH), Germany
Vodafone – Panafon Anonymi Elliniki Etaireia Tilepikoinonion, Greece
Software Mind SP. Z O.O., Poland
The University of Sheffield, United Kingdom
Motorola Limited, United Kingdom
Sheffield City Council, United Kingdom

Administrative details
WeKnowIt (ICT-215453) is a Integrated Project (IP) of the European Union's 7th Framework Programme: Information and Communication Technologies (ICT) – Call 1.
The project started on 1 April 2008, and will finish on 31 March 2011 (36 months).
There are 10 participants from 5 countries involved in the project, and the EC contribution is 5.61 million Euros (total cost: €8.05m).



This page is maintained by: The INFSO.E2 Webmaster
Last updated on: 2009-10-30
Top | CORDIS Services | Help Desk | DG Information Society & Media

CORDIS is managed by the Publications Office