| Knowledge about which competitor company has invested how much money in a specific media campaign is very important for the highest management level of companies. Such information is gathered through global advertisement expenditure measurement, which is performed by media monitoring companies. This type of business intelligence is a very complex task, which is currently performed manually and therefore is very expensive. A media campaign is defined as the universe of measures in order to fulfil a specific objective. MediaCampaign's scope is on discovering, inter-relating and navigating cross-media campaign knowledge and to automate a large degree of the detection and tracking of media campaigns on television, Internet and in the press. For the pilot system developed within the project the project focus is on a concrete example for a media campaign: advertisement campaigns. However, the approach taken and the component implementations will allow easy setup of a system for monitoring and analysis of other campaigns (e.g. political and social ones). In order to get the information as fast as possible the currently manual process to acquire the necessary data will be significantly accelerated by the MediaCampaign framework. Scientifically there are a number of high-risk research objectives to meet: (1) creation of a knowledge model for semantic description of media campaigns in general, (2) identification & tracking of new media campaigns in different media and (3) modelling of domain specific ontologies which relate media campaigns over different media and countries. In coherence with these scientific objectives R&D work carried out within MediaCampaign includes the design and implementation of a specific media campaign ontology (media presence & campaign ontology MEPCO), a module for cross relation of specific campaigns, algorithms for detecting advertisements over different media (utilizing different modalities), audio analysis algorithms and the detection plus tracking of new campaigns. In addition the R&D work will be concentrated on exploitation of semantic web technologies with a focus on querying and navigating within multi dimensional information spaces. In connection with cross relation a unique innovative approach will be followed by combining formal modelling of media campaigns with semantic content consolidation and reasoning and large-scale knowledge bases. With regard to audio analysis R&D work will be concentrated on segmentation, word spotting, dedicated speech to text modules and jingle recognition. In order to relate media campaigns over the different media TV, press and Internet we adopt the following innovative algorithm: use speech transcripts and OCR techniques to receive (noisy) text from A/V content, apply terminology identification and unsupervised clustering techniques for decision of most significant text parts, use those parts for a focussed web search and process the results of this search with IE methods in order to spot whether the campaign is new or not. |
