I am a postdoc researcher at the DISCo (Department of Informatics, Systems and Communication of University of Milano – Bicocca) since 2017. I am a member of INSID&S Lab. During my academic life, I have established a
collaboration with the DWS Group at the University of Mannheim, Germany, where I spent my 6-months period abroad and the University of Shanghai, China.
My research activities focus on overcoming challenges related to Profiling Linked Data. The profile of a data is fundamental for the decision-making process and is the basic requirement towards the dataset understanding. During my research I have covered three aspect of profiling: topic-based, schema-based and linkage-based. For the topic-based profiling, an approach to automatically classify datasets in one of the categories used in the Linked Data cloud is proposed. ABSTAT is a schema-based profiling tool that provides an overview about the relations in the data. ABSTAT’ profiles are concise and informative enough to summarize the whole data. Many datasets in the LD cloud contain similar information for the same entity. In order to fully exploit its potential LD datasets should made this information explicit. Linkage-based profiling provides information about the number of equivalent entities between datasets thus revealing possible errors.
Currently I am working on quality assessment and error detection techniques by applying constraints using SHACL and machine learning algorithms.
COMSODE – Components Supporting the Open Data Exploitation (2013-2015) – EU FP7
euBusinessGraph – euBusinessGraph aims to create a crossborder knowledge graph of companies, and a set of innovative business products and services build upon the knowledge graph (2017-2019) – H2020
spahiu at disco dot unimib dot it