My research focuses on Big Data analysis, summarization and user-friendly visualization. In particular, I develop data clustering algorithms for fully automatic construction of user-friendly taxonomies of topics for various datasets including scientific papers. Also, I develop web components for visual exploration of the constructed taxonomies.
My work has been supported by the Swiss National Science Foundation, European Research Council and by the University of Fribourg.
Clustering, community structure discovery, semantic web, ontology building, human-computer interaction, data visualization, clustering benchmarking, clustering quality measurement.
- Research Assistant in UNIFR (2014-nowadays)
- Startups (2012-nowadays)
- Lead/Senior Engineer & Project Leader in Samsung Ukraine R&D Center (2009-2012)
- System Developer in Symantec / PC Tools (2008-2009)
- QA and System Developer in Zoral Labs (2006-2008)
- MSc in Computer Science, specialization: Computer Systems and Networks. NTUU KPI, Ukraine (2005-2007)
- BSc in Computer Science, specialization: Computer Engineering. NTUU KPI, Ukraine (2001-2005)
Projects at XI-Lab
While at UNIFR I implemented or contributed to the projects including:
- DAOC: Accurate and Parameter-free Clustering of Large Networks
- Clubmark: a parallel isolation framework for benchmarking and profiling clustering algorithms
- PyExPool: Python Multi-Process Execution Pool, a concurrent asynchronous execution pool with custom resource constraints (memory, timeouts, affinity, CPU cores and caching), load balancing and profiling capabilities of the external apps on NUMA architecture
- xmeasures: extremely fast evaluation of the extrinsic clustering measures: (mean) F1 measures family and Omega Index (Fuzzy Adjusted Rand Index) for the multi-resolution clustering with overlaps, standard NMI, clusters labeling
- GenConvNMI: Generalized Conventional Mutual Information
- LFR-Benchmark_UndirWeightOvp: Extended version of the Lancichinetti-Fortunato-Radicchi Benchmark for Undirected Weighted Overlapping networks to evaluate clustering algorithms using generated ground-truth communities
- PyCaBeM: Python Benchmarking Framework for the Clustering Algorithms Evaluation
- HiReCS: parameter-free deterministic hierarchical clustering library
- ScienceWISE: a semantic layer on top of arXiv.org to browse and discover research papers
- Artem Lutov, Soheil Roshankish, Mourad Khayati, and Philippe Cudré-Mauroux. “StaTIX — Statistical Type Inference on Linked Data.” In Proceedings of the IEEE BigData 2018 - 4th Special Session on Intelligent Data Mining, 2018. Bibtex PDF
- Artem Lutov, Mourad Khayati, and Philippe Cudré-Mauroux. “Clubmark: a Parallel Isolation Framework for Benchmarking and Profiling Clustering Algorithms on NUMA Architectures.” In Proceedings of the 2018 IEEE International Conference on Data Mining (ICDM), 1481–86, 2018. Bibtex Slides PDF
- Andrea Martini, Artem Lutov, Valerio Gemmetto, Andrii Magalich, Alessio Cardillo, Alex Constantin, Vasyl Palchykov, et al. “ScienceWISE: Topic Modeling over Scientific Literature Networks.” ArXiv e-Prints, 2016. Bibtex PDF