Modern world is built around knowledge and it is impossible to think without knowledge.
As the volume of online information continues to grow, the need for innovation in the domain of knowledge extraction is getting more and more pressing.
This is equally important for general-purpose sources as well as for domain-specific information, e.g., for physics or computer science.
While there are many solutions providing effective information management for structured, domain-specific data, large amounts of information still remain in unstructured form (e.g., text) and are hence difficult to process.
My research is focused on developing new methods providing effective knowledge extraction, processing and discovery for unstructured technical sources.
Specifically, my focus is on performing named-entity recognition and disambiguation for scientific documents, establishing semantic relationships between scientific entities, and providing effective document discovery based on those entities.
My full CV is available on: careers.stackoverflow.com/prokofyev
Semantic Web; Topic modeling; Text classification; Word Sense Disambiguation
- Roman Prokofyev, Djellel Difallah, Michael Luggen, and Philippe Cudré-Mauroux. “SwissLink: High-Precision, Context-Free Entity Linking Exploiting Unambiguous Labels.” In Proceedings of the 13th International Conference on Semantic Systems (SEMANTICS2017). Amsterdam, The Netherlands, 2017. Bibtex Slides PDF
- Alberto Tonon, Michele Catasta, Roman Prokofyev, Gianluca Demartini, Karl Aberer, and Philippe Cudré-Mauroux. “Contextualized Ranking of Entity Types Based on Knowledge Graphs .” Web Semantics: Science, Services and Agents on the World Wide Web , 2016, - . Bibtex PDF
- Julia Proskurnia, Ruslan Mavlyutov, Roman Prokofyev, Karl Aberer, and Philippe Cudré-Mauroux. “Analyzing Large-Scale Public Campaigns on Twitter.” In International Conference on Social Informatics, 225–43. Springer, 2016. Bibtex
- Roman Prokofyev, Alberto Tonon, Michael Luggen, Loic Vouilloz, Djellel Eddine Difallah, and Philippe Cudré-Mauroux. “SANAPHOR: Ontology-Based Coreference Resolution.” In The Semantic Web - ISWC 2015 - 14th International Semantic Web Conference,
Bethlehem, PA, USA, October 11-15, 2015, Proceedings, Part I, 458–73, 2015. Bibtex Slides PDF
- Roman Prokofyev, Gianluca Demartini, and Philippe Cudré-Mauroux. “Effective Named Entity Recognition for Idiosyncratic Web Collections.” In WWW. Seoul, Korea: IW3C, 2014. Bibtex PDF
- Roman Prokofyev, Ruslan Mavlyutov, Martin Grund, Gianluca Demartini, and Philippe Cudré-Mauroux. “Correct Me If I’m Wrong: Fixing Grammatical Errors by Preposition Ranking.” In CIKM. CIKM ’14. Shanghai, China: ACM, 2014. Bibtex PDF
- Roman Prokofyev, Gianluca Demartini, Alexey Boyarsky, Oleg Ruchayskiy, and Philippe Cudré-Mauroux. “Ontology-Based Word Sense Disambiguation for Scientific Literature.” In ECIR, 594–605. Moscow, 2013. Bibtex PDF
- Roman Prokofyev, Alexey Boyarsky, Oleg Ruchayskiy, Karl Aberer, Gianluca Demartini, and Philippe Cudré-Mauroux. “Tag Recommendation for Large-Scale Ontology-Based Information Systems.” In International Semantic Web Conference (2), 325–36. Springer, 2012. Bibtex PDF
Talks & Presentations
14.10.2015 SANAPHOR @ ISWC 2015
04.11.2015 “Correct Me If I’m Wrong @ CIKM 2014