FROCKG (Fact Checking for Large Enterprise Knowledge Graphs) is a Eurostars-2 project that aims to develop accurate & time-efficient approaches to quantify and provide evidence for the veracity of facts contained in enterprise knowledge graphs. The approach combines machine learning, large-scale incremental indexes, efficient computations of stationary distributions and consistency checks to quantify how true/untrue facts are likely to be. The project delivers a prototype, which is integrated into commercial solutions and deployed in three use cases (Finance, Pharma, Cultural Heritage, Linked Open Data).
Enterprise knowledge graphs underpin business-critical decisions. E.g., facts extracted from news streams on business entities influence financial markets and business decisions by inducing 280% higher trade volume and 180% higher price change within 10 minutes of availability (Fedyk, 2018; see Front Page News:The Effect of News Positioning on Financial Markets). The results of the FROCKG project enable companies to ensure the veracity of facts obtained from their data suppliers before taking business-critical decisions based on them.
The goal of the FROCKG project is the development of a framework that allows (1) the quantification of the veracity of facts (i.e., their likelihood to be true/false) found in enterprise knowledge graphs (EKG) and (2) the collection of evidence for this veracity. EKG are increasingly business-critical to a growing number of modern knowledge-driven companies. Determining the veracity of the facts in these EKG is hence mission-critical. The project is motivated by the current lack of market-ready (w.r.t. cost and scalability) solutions for this purpose.
Sirma AI, trading as Ontotext, is part of Sirma Group Holding and provides a commercial suite of semantic technologies enabling better content management, knowledge discovery and semantic search.
Ontotext is responsible for developing the tools for incremental knowledge extraction from both structured and unstructured data sources. As part of the FROCKG project, Ontotext is building a knowledge graph, integrating several structured datasets from the life science and pharmaceutical domain. In order to enrich the data in the knowledge graph, Ontotext is also developing text analysis pipelines for extracting knowledge from unstructured data. The platform allows utilizing fact-checking algorithms, developed by Paderborn University for determining the veracity of facts in the knowledge graph as well as manual editing and curation of facts, using tools developed by metaphacts. The automatically validated and curated facts then are continuously integrated into the knowledge graph.
Ontotext is also developing one of the three use cases, which would be used in the FROCKG project (along with Cultural Heritage and Linked Open Data) – Pharma fake news detection, which aims at analyzing how data in the Life Science and Pharmaceutical domain could be used to manipulate the Financial markets.
The FROCKG project has received funding from the Eurostars-2 joint programme with co-funding from the European Union H2020 research and innovation programme under contract No E-1/06.04.2020 with local funding agency (BSMEPA).