In conjunction with the 11th Extended Semantic Web Conference (ESWC 2014)
Knowledge discovery is a well-established field with a large community investigating methods for the discovery of patterns and regularities in large data sets, including relational databases and unstructured text. Research in this field has led to the development of practically relevant and scalable approaches such as association rule mining, subgroup discovery, graph mining, and clustering.
At the same time, the Web of Data has grown to one of the largest publicly available collections of structured, cross-domain data sets. While the growing success of Linked Data and its use in applications, e.g., in the e-Government area, has provided numerous novel opportunities, its scale and heterogeneity are posing several challenges to the field of knowledge discovery and data mining:
- The extraction and discovery of knowledge from very large data sets;
- The maintenance of high quality data and provenance information;
- The scalability of processing and mining the distributed Web of Data; and
- The discovery of novel links, both on the instance and the schema level.
Contributions from the knowledge discovery field may help foster the future growth of Linked Open Data. Some recent works on statistical schema induction, mapping, and link mining have already shown that there is a fruitful intersection of both fields. With the proposed workshop, we want to investigate possible synergies between both the Linked Data community and the field of Knowledge Discovery, and to explore novel directions for mutual research. On the one hand, we wish to stimulate a discussion about how state-of-the-art algorithms for knowledge discovery and data mining can be adapted to fit the characteristics of Linked Data, such as its distributed nature, incompleteness (i.e., absence of negative examples), and identify concrete use cases and applications. On the other hand, we hope to show that Linked Data can support traditional knowledge discovery tasks (e.g., as a source of additional background knowledge and of predictive features) for mining from existing, not natively linked data like, for instance, in business intelligence settings.
Authors of contributed papers are especially encouraged to publish their data sets and/or the implementation of their algorithms, and to discuss these implementations and data sets with other attendees. The goal is to establish a common benchmark that can be used for competitive evaluations of algorithms and tools.
This workshop will join two successful series of past events. It follows the 2012 and 2013 editions of Know@LOD, held at ESWC, as well as the Data Mining on Linked Data (DMoLD) workshop, which was held at ECML/PKDD 2013. Besides a track for research papers, the workshop will host the second Linked Data Mining Challenge (the first having been at DMoLD).
The workshop is kindly supported by www.KDnuggets.com: Analytics, Big Data, Data Mining, & Data Science Resources.