Invited Talk

Title: Knowledge Graphs and Natural Language Processing: a symbiosis


In this talk, we explore the symbiotic relationship between knowledge graphs (KGs) and natural language processing (NLP). We start by investigating in how far information extraction as an important task in the field of NLP can profit from the knowledge encoded in knowledge graphs, DBpedia in particular. We then introduce the important concept of scientific knowledge graphs, referring to knowledge graphs that capture the body of knowledge of a certain scientific discipline. We argue why such scientific knowledge graphs will increasingly play a role in the near future. We show in particular how NLP techniques can be used to populate scientific knowledge graphs, in particular in the domain of medicine. We present results of the PSINK project, the goal of which is to automatically construct a large knowledge graph comprising of all the pre-clinical study results in the domain of spinal cord injuries. We close by showing some powerful applications that can be built on top of scientific knowledge graphs.

Philipp CimianoPhilipp Cimiano is the head of the Semantic Computing Group at Bielefeld University. He is also affiliated with the Cognitive Interaction Technology Excellence Cluster (CITEC). Before joining the Bielefeld University, he was an assistant professor at Delft University of Technology (2008-2009) and a postdoctoral researcher at the Institute AIFB of Karlsruhe University (TH). Philipp Cimiano is mainly interested in topics at the intersection between knowledge representation and text processing including: text mining, computational semantics, information retrieval, question answering, ontology learning, ontology localization, etc. He is editorial board member of the Semantic Web Journal.


Wednesday November 25th – 10:30-12:30 CET

Link to the meeting (WebEx)
The password will be sent on Slack

10:30-11:00 Welcome to NL4AI 2020

11:00-12:00 Invited Talk: Philipp Cimiano [Slides (PDF)]

12:00-12:30 Discussion

Thursday November 26th – 14:00-16:00 CET

Link to the meeting (WebEx)
The password will be sent on Slack

14:00-15:00 Question time with the authors

  • Pierluigi Cassotti, Pierpaolo Basile, Marco de Gemmis and Giovanni Semeraro
    A comparative study of approaches for the diachronic analysis of the Italian language
    [Paper (PDF)] [Video]

  • Dagmar Monett, Niklas Lampe, Max Ehrlicher-Schmidt and Nikolaj Bewer
    Intelligence Catalog-guided Tracking of the Evolution of (machine) Intelligence: Preliminary results
    [Paper (PDF)] [Video]

  • Tahir Mehmood, Ivan Serina, Alberto Lavelli and Alfonso Gerevini
    Knowledge Distillation Techniques for Biomedical Named Entity Recognition
    [Paper (PDF)] [Video]

  • Nhu Khoa Nguyen, Emanuela Boros, Gaël Lejeune and Antoine Doucet
    Impact Analysis of Document Digitization on Event Extraction
    [Paper (PDF)] [Video]

  • Roberto Salvaneschi, Daniela Muradore, Andrea Stanchi and Viviana Mascardi
    FrEX: Extracting Property Expropriation Frame Entities from Real Cases
    [Paper (PDF)] [Video]

  • Jérôme Delobelle, Amaury Delamaire, Elena Cabrio, Ramon Ruti and Serena Villata
    Sifting the Arguments in Fake News to Boost a Disinformation Analysis Tool
    [Paper (PDF)] [Video]

  • Lucia C. Passaro and Alessandro Lenci
    Less is MORE: a MultimOdal system for tag REfinement
    [Paper (PDF)] [Video]

15:00-16:00 Discussion panel: Natural Language Processing and Explainablillty of AI systems: an interwined story. Panelists: Serena Villata, Cataldo Musto, Danilo Croce.

Friday November 27th – 9:00-10:00 CET

9:00-9:50 Question time with the authors

  • Ginevra Carbone and Gabriele Sarti – ETC-NLG: End-to-end Topic-Conditioned Natural Language Generation
    [Paper (PDF)] [Video]

  • Claudio Greco, Alberto Testoni and Raffaella Bernardi
    Which turn do neural models exploit the most to solve GuessWhat? Diving into the dialogue history encoding in Transformers and LSTMs
    [Paper (PDF)] [Video]

  • Matteo Zubani, Ivan Serina and Alfonso Emilio Gerevini
    A Comparison of Natural Language Understanding Services to build a chatbot in Italian
    [Paper (PDF)] [Video]

  • Giancarlo Xompero, Cristina Giannone, Fabio Massimo Zanzotto, Andrea Favalli and Raniero Romagnoli
    Injecting Designers’ Knowledge in Conversational Neural Network Systems
    [Paper (PDF)] [Video]

  • Carmelo Fabio Longo and Corrado Santoro
    AD-CASPAR: Abductive-Deductive Cognitive Architecture based on Natural Language and First Order Logic Reasoning
    [Paper (PDF)] [Video]

  • Yue Zhan and Michael Hsiao
    Breaking Down High-level Robot Path-Finding Abstractions in Natural Language Programming
    [Paper (PDF)] [Video]

9:50-10:00 Goodbye