Jeroen De Knijf

BioGraph: unsupervised biomedical knowledge discovery via automated hypothesis generation

We present BioGraph, a data integration and data mining platform for the exploration and discovery of biomedical information. The platform offers prioritizations of putative disease genes, supported by functional hypotheses. We show that BioGraph can retrospectively confirm recently discovered disease genes and identify potential susceptibility genes, outperforming existing technologies, without requiring prior domain knowledge. Additionally, BioGraph allows for generic biomedical applications beyond gene discovery. BioGraph is accessible at http://www.biograph.be.

FAT-CAT: Frequent Attributes Tree Based Classification

The natural representation of XML data is to use the underlying tree structure of the data. When analyzing these trees we are ensured that no structural information is lost. These tree structures can be efficiently analyzed due to the existence of frequent pattern mining algorithms that works directly on tree structured data. In this work we describe a classification method for XML data based on frequent attribute trees. From these frequent patterns we select so called emerging patterns, and use these as binary features in a decision tree algorithm. The experimental results show that combining emerging attribute tree patterns with standard classification methods, is a promising combination to tackle the classification of XML documents.

GaMuSo: graph base music recommendation in a social bookmarking service

In this work we describe a recommendation system based upon user-generated description (tags) of content. In particular, we describe an experimental system (GaMuSo) that consists of more than 140.000 user-defined tags for over 400.000 artists. From this data we constructed a bipartite graph, linking artists via tags to other artists. On the resulting graph we compute related artists for an initial artist of interest. In this work we describe and analyse our system and show that a straightforward recommendation approach leads to related concepts that are overly general, that is, concepts that are related to almost every other concept in the graph. Additionally, we describe a method to provide functional hypothesis for recommendations, given the user insight why concepts are related. GaMuSo is implemented as a webservice and available at: music.biograph.be.