I. Kivrakidis, E. Rigas, N. Bassiliades, “Towards Enriching the Electric Vehicle Knowledge Graph by Linking it to DBpedia”, AIKGC2024: AI-Driven Knowledge Graph Construction, Workshop at SETN 2024, Sep 11-13, 2024, Athens, Greece, Proc. of the 13th Hellenic Conference on Artificial Intelligence (SETN '24), Association for Computing Machinery, New York, NY, USA, Article 38, 1-4.
The automotive industry is focusing on Electric Vehicles (EVs) for their efficiency in reducing oil consumption and emissions. However, the EV market’s diversity in battery capacities, classifications, and connector types creates a lack of standardization. Researchers are exploring advanced data and knowledge management methods, with Knowledge Graphs (KGs) emerging as a promising solution. KGs represent data in a way that reflects human understanding, promoting natural human-machine interactions and enhancing AI’s insights. Their structure and constraints are defined by a vocabulary or ontology. This paper presents ongoing work towards enriching an existing Electric Vehicle Knowledge Graph (EVKG), which is specified by the Electric Vehicle Ontology (EVO). To achieve this, we utilized the Python programming language to create labels for each resource in the EVKG. These labels were then used to match and link these resources to their corresponding entries in DBpedia through the use of the owl:sameAs and rdfs:seeAlso properties. To ensure the matching was as accurate as possible, we developed two algorithms: one employing string matching and the other using word vectorization and distance techniques.
