The results of the 1st year of implementation of the LinDA* FP7-610565 project have been successfully demonstrated during the periodic project review on 29/1 at the European Commission in Luxembourg. Among these, the first release of the LinDA workbench, which is now ready for deployment into the LinDA Pilots, as well as the setup of the pilots (fostering Business Intelligence Analytics, Environmental Analytics, Media Analytics) that are going to be executed and validated during the second year of the project.
UBITECH’s contribution to the project regards mainly the design and development of the Linked Data Analytics and Data Mining Component of the LinDA workbench as well as the leadership of WP4 that regards the LinDA pilots. In more detail, the Linked Data Analytics and Data Mining Component supports the realisation of analysis based on the consumption and production of Linked Data. A library of basic and robust data analytic functionality is provided through the support of a set of algorithms, enabling SMEs to utilise and share analytic methods on Linked Data for the discovery and communication of meaningful new patterns that were unattainable or hidden in the previous isolated data structures. Specifically, integration of the Weka open-source tool and the R open-source project for statistical computing is realised, while the following algorithms are supported per category: (i) Classification Analysis Algorithms (J48, M5P), (ii) Association Analysis Algorithms (Apriori), (iii) Statistics/Forecasting Analysis Algorithms (Linear Regression, Multiple Linear Regression, Arima), (iv) Geospatial Analysis Algorithms (Morans I, Kriging, NCF correlogram), and (v) Clustering Algorithms (KMeans Partitioning, Ward Hierarchical Agglomerative, Model Based Clustering). High priority is given to the user friendliness of the provided interfaces based on the design of specialized workflows per algorithm category (e.g. workflows for supervised learning techniques such as classification and regression/forecasting algorithms and unsupervised learning techniques such as clustering and pattern discovery (association) algorithms).
For more information, please visit the project’s website or contact us.
* The LinDA project addresses one of the most significant challenges of the usage and publication of Linked Data, the renovation and conversion of existing data formats into structures that support the semantic enrichment and interlinking of data. The set of tools provided by LinDA will assist enterprises, especially SMEs which often cannot afford the development and maintenance of dedicated information analysis and management departments, in efficiently developing novel data analytical services that are linked to the available public data therefore contributing to improve their competitiveness and stimulating the emergence of innovative business models.
