Ontology-driven Event Type Classification in Images

Abstract

Event classification can add valuable information for semantic search and the increasingly important topic of fact validation in news. So far, only few approaches address image classification for newsworthy event types such as natural disasters, sports events, or elections. Previous work distinguishes only between a limited number of event types and relies on rather small datasets for training. In this paper, we present a novel ontology-driven approach for the classification of event types in images. We leverage a large number of real-world news events to pursue two objectives: First, we create an ontology based on Wikidata comprising the majority of event types. Second, we introduce a novel large-scale dataset that was acquired through Web crawling. Several baselines are proposed including an ontologydriven learning approach that aims to exploit structured information of a knowledge graph to learn relevant event relations using deep neural networks. Experimental results on existing as well as novel benchmark datasets demonstrate the superiority of the proposed ontology-driven approach.

Graph Based Analytics and Review of Knowledge Graph Field by Using Two Decade Data- Finding Paradigms and Opportunities

“…The Knowledge Graph (KG) is a graph-based representation of knowledge 36) , where nodes of the graph represent entities of interest and edges represent relationships 37,38,40) . Entities are titled with attributes and edges capture relationships among them.…”

Understanding image-text relations and news values for multimodal news analysis

“…Previous works mix concrete and abstract concepts under one relation, such as “text is represented” relation in Vempala and Preotiuc-Pietro ( 2019 ), “contextual” relation in Kruk et al ( 2019 ), and Otto et al ( 2019b ) referring to entities as “main objects” in foreground of an image. From the computational perspective, object (Deng et al, 2009 ; Thomee et al, 2016 ), scene (Xiao et al, 2010 ; Zhou et al, 2018 ), event (Xiong et al, 2015 ; Müller-Budack et al, 2021a ), and action recognition (Heilbron et al, 2015 ; Gu et al, 2018 ) models and their classes provide the list of all concepts and entities that have clear depictions and are thus identifiable in an image.…”

OEKG: The Open Event Knowledge Graph

“…Event classification in images is an important task for various applications in the fields of computer vision, including geolocation estimation and place classification [17]. Such tasks typically rely on the existence of a well-defined class hierarchy and the availability of images.…”