Prestigious Women Researchers Lecture Series | Anastasia Dimou : 'Understanding, Semantically Annotating and Validating Tables'

In this talk we will discuss the methods which are followed so far to (semi-)automatically annotate tables and its relationship with declarative methods to create knowledge graphs from semi-structured data. The talk will cover the latest research of our team on automated the annotation of tables considering existing knowledge graphs, such as Wikidata, as well as language models which won the award for this year’s SemTab challenge (a challenge that compares systems which automatically understand and annotate tables). We will discuss how our automated solution can be combined with declarative methods for the construction and validation of knowledge graphs in the form of RDF gaphs, as well as how we plan to leverage this solution to enrich a knowledge graph about machine learning algorithms.

About me (Anastasia Dimou)

Anastasia Dimou is tenure track assistant professor on Machine Learning and Artificial Intelligence for cost- and data-efficient solutions since September 2021 at the Declarative Languages and Artificial Intelligence (DTAI) unit of the Computer Science Deartment of KU Leuven where she currently leads a team of 8 members. Before she was scientific researcher at imec (2013-2021). She received her PhD in 2017 from Antwerp University pertaining to high quality knowledge graph construction from heterogeneous data for which she won the SWSA distinguished dissertation award. Her research interests focus on declarative and efficient knowledge graph construction from heterogeneous data; validation, quality assessment and benchmarking of knowledge graphs; (semi-)automated knowledge graph construction with human-in-the-loop, data spaces, as well as on the fusion of Semantic Web technologies & Machine Learning.

Since 2013, Prof. Dimou was involved in around 20 Flemish, Belgian and European industry-driven and fundamental research projects. She is the promotor of the Belgian Network on Knowledge Graph for Data Integration (KG4DI, FWO WoG, w3id.org/ kg4di/) which brings together 7 research labs of Belgium working on knowledge graphs and semantic web technologies, as well as 10 labs from abroad. She is also the chair of the W3C Community Group on Knowledge Graph Construction (W3C-KGC, www.w3.org/community/kg-construct/), with more than 160 members from all over the world and the science communication officer of the Distributed Knowledge Graphs COST Action (DKG, cost-dkg.eu/) with more than 100 members from more than 30 countries in Europe.