Petar Ristoski's Workflows

Using Graph Kernels for Feature Generation... (1)

• Download

This example shows how to use graph kernels for feature generation.In this example we use the Root RDF Walk Count Kernel, and the Fast RDF WL Sub Tree Kernel.The input data for the process can be found here.More information about the process can be found here.

Created: 2015-05-04 | Last updated: 2015-05-04

Hybrid Recommender System Using Linked Op... (1)

• Download

This process is using the Rapid Miner Linked Open Data extension and the Recommender extension, to build a hybrid Linked Open Data enabled recommender system for books.The input data for the process can be found here.More information about the process can be found here.

Created: 2014-05-15 | Last updated: 2014-05-15

Predicting the fuel consumption of cars (1)

• Download

This process is using features extracted from DBpedia to predict the fuel consumption of cars. It uses operators from the Linked Open Data and the Weka extensions.The process first reads a list of cars with the fuel consumption value. The DBpedia lookup linker is used to link the car names to DBpeida resources. The established links are used to generate additional features, i.e. direct types and categories of each car. Finally, a M5 Rules Operator is used to predict the fuel co...

Created: 2014-05-14 | Last updated: 2014-05-15