The results of the second year of implementation of the LinDA* FP7-610565 project have been successfully demonstrated during the final project review on January 21st, 2016 at the European Commission in Luxembourg. These results include the final release of the LinDA workbench along with the performance evaluation of the LinDA tools, as well as the results from the operation of the LinDA pilots (fostering Business Intelligence Analytics, Environmental Analytics, Media Analytics) along with a set of suggestions for LinDA adopters.
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 R open-source project for statistical computing and the Weka open-source tool 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). UBITECH has also been actively involved in the design and operation of all the pilots and especially the Environmental Analytics and Media Analytics pilots. In the Environmental Analytics pilot, linked data technologies are exploited for examining the health impact of outdoor air pollution in international, national and regional level, while in the Media Analytics pilot, linked data are being used towards the realisation of data-driven investigative journalism as well as the production and consumption of linked data media analytics.
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.
