Program
Committee Knowledge Graphs have been used in various machine learning tasks by deriving latent feature representations of entities and relations. Knowledge Graphs represent formal semantics by describing entities and relationships between them, and can use ontologies as a schema layer of reference. This way, it is possible to retrieve implicit knowledge through logical inference rather than only allowing queries that request explicit knowledge. Deep Learning methods have emerged from machine learning approaches and became essential for the resolution of several tasks within the arti cial intelligence spectrum. Recently, Deep Learning methods have been used in conjunction with Knowledge Graphs (i.e., to represent relationship of the graph in a vector space, to allow companies nd patterns in real-time between interconnected entities, to keep track of inventories of parts further allowing nding materials used in di erent products, etc.). Therefore, it has become critical that the Deep Learning and Knowledge Graphs communities join their forces in order to develop more e ective algorithms and applications. This workshop aimed at reinforcing the relationships between these communities and intended to be at the center of shared works around topics such as Deep Learning, Knowledge Graphs, Natural Language Processing, Computational Linguistics, Big Data, and so on.
Therefore, the goal of this workshop was to provide a meeting forum where discussions between the relevant stakeholders (researchers from academia, industry, and businessmen) could be stimulated within the Deep Learning and Knowledge Graphs domains. As the previous edition, this year we noticed a general attention to our workshop given that more than 11 submissions we received and the high number of participants we noticed during the workshop day. Ten papers have been accepted and discussed within the workshop by authors from di erent international institutions. They covered topics such as question answering, temporal Knowledge Graph embeddings, transformer-based entity detection, language model detection, link prediction, sparsity in cultural heritage image archives, multi-label classi cation. We had as invited speaker Prof. Afshin Sadeghi who discussed how to embed a type of dynamic KGs that constantly grow by integrating a stream of new facts. These ever-growing graphs are known as Accrescent knowledge Graphs (AKG). In contrast to discrete-time dynamic graphs that di erent snapshots of a KG are considered, the training of AKGs involves training upon the stream of new triples. We also thank the program committee for their time and work for reviewing the submitted papers. Although the workshop was held remotely due to the COVID-19 pandemic, it has been successful and attended by more than 60 participants from all around the world. On the workshop website6 it is possible to see screenshots re ecting some moments of the workshop.
6 https://alammehwish.github.io/dl4kg2022/ Towards A Question Answering System over Temporal Knowledge Graph Embedding Kristian Otte, Kristian Simoni Vestermark, KHuan Li and Daniele Dell'Aglio Transformer-based Subject Entity Detection in Wikipedia Listings, Nicolas Heist and Heiko Paulheim Improving Language Model Predictions via Prompts Enriched with Knowledge Graphs, Ryan Brate, Minh-Hoang Dang, Fabian Hoppe, Yuan He, Albert Meron~o-Pen~uela and Vijay Sadashivaiah Knowledge Graph Embeddings for Link Prediction: Beware of Semantics!, Nicolas Hubert, Pierre Monnin, Armelle Brun and Davy Monticolo Neuro-symbolic learning for dealing with sparsity in cultural heritage image archives: an empirical journey, Agnese Chiatti and Enrico Daga Bilingual Question Answering over DBpedia Abstracts through Machine Translation and BERT Michalis Mountantonakis,, Michalis Bastakis, Loukas Mertzanis and Yannis Tzitzikas A Closer Look at Sum-based Embeddings for Knowledge Graphs Containing Procedural Knowledge, Richard Nordsieck, Michael Heider, Anton Hummel and Joerg Haehner Knowledge Graph Embeddings for Causal Relation Prediction, Aamod Khatiwada, Sola Shirai, Kavitha Srinivas and Oktie Hassanzadeh Multi-label Classi cation using BERT and Knowledge Graphs with a Limited Training Dataset, Malick Ebiele, Lucy McKenna, Malika Bendechache and Rob Brennan