CKDD 2012 - 3rd INTERNATIONAL TRACK ON COOPERATIVE KNOWLEDGE DISCOVERY & DATA MINING

Knowledge Discovery and Data Mining require users skills at various levels of data processing and of their application domains. Collaboration offers the potential of improved results by harnessing dispersed expertise and enabling knowledge sharing and learning. In Knowledge Discovery, collaboration is also motivated by different expertise and tools the users may have, as well as by the distribution of the collected datasets for mining and analysis. Numerous organizations start collecting data in geographically distributed heterogeneous repositories and perform decision making in a distributed fashion. Extracting useful knowledge from datasets produced by different parties presents new challenges, such as quality, heterogeneity, privacy, ownership, remote access and/or processing, availability of resources. These can find a technological solution in collaborative environments for distributed Data Mining.
Moreover, there have been significant technological advances in both hardware and software for collaborative and distributed environments in the last decade or so. These systems are quite mature and are semantic technologies for integration and interoperability of information sources, application interoperability, Computer Supported Cooperative Work environments, and Grid infrastructures. They can be of great help in very large complex data mining applications: for instance, distributed Data Mining algorithms can help manage distributed heterogeneous data as well as deal with partial information and models. In addition, collaboration tools and platforms can help different geographically dispersed groups to work together on the same data-driven applications in a collaborative manner. Then, semantic technologies can help dataset understanding, integration, consolidation, and composition. There have been various projects in collaborative data analysis and the distributed computation aspects of the data mining techniques.