Mobile link prediction: Automated creation and crowdsourced validation of knowledge graphs

https://doi.org/10.1016/j.micpro.2021.104335Get rights and content
Under a Creative Commons license
open access

Abstract

Building trustworthy knowledge graphs for cyber–physical social systems (CPSS) is a challenge. In particular, current approaches relying on human experts have limited scalability, while automated approaches are often not validated by users resulting in knowledge graphs of questionable quality. This paper introduces a novel pervasive knowledge graph builder for mobile devices that brings together automation, experts’ and crowdsourced citizens’ knowledge. The knowledge graph grows via automated link predictions using genetic programming that are validated by humans for improving transparency and calibrating accuracy. The knowledge graph builder is designed for pervasive devices such as smartphones and preserves privacy by localizing all computations. The accuracy, practicality, and usability of the knowledge graph builder is evaluated in a real-world social experiment that involves a smartphone implementation and a Smart City application scenario. The proposed methodology of knowledge graph building outperforms a baseline method in terms of accuracy while demonstrating its efficient calculations on smartphones and the feasibility of the pervasive human supervision process in terms of high interactions throughput. These findings promise new opportunities to crowdsource and operate pervasive reasoning systems for cyber–physical social systems in Smart Cities.

Keywords

Knowledge graph
Ontology
Cyber–physical-social system
Link prediction
Genetic programming
Crowdsourcing

Cited by (0)

Mark C. Ballandies is a Ph.D. candidate at the professorship of Computational Social Science (Prof. Dirk Helbing) at ETH Zurich, Switzerland, where he also received his B.Sc. and M.Sc. in computational science and engineering. He gained expertise in the fields of evolutionary algorithms, machine learning, and software development in numerous scientific and industry works. He is an alumnus of the Foundation of German Business, president of the NGO Education Matters e.V., treasurer of the Club Alpbach Zurich, member of the dcentgroup and initiator of the smarte Dörfer association. He is also the founder of the netti. app, a knowledge graph editor that won as part of the EmpowerPolis project the 1st prize at the ETH Policy Challenge. He has been awarded a prestigious fellowship at the ETH library lab. He has been involved in EU projects such as FuturICT 2.0 and has supervised several Bachelor and Master students while he contributed to an innovative course in the area of blockchain and cryptoeconomic systems.

Marks research interests lie in multi-dimensional incentive systems for information sharing, where he combines the fields of semantic networks, ethical data sharing and cryptoeconomics.

Evangelos Pournaras is an Associate Professor at Distributed Systems and Services group, School of Computing, University of Leeds, UK. He is also currently research fellow in blockchain industry. He has more than 5 years of experience as senior scientist and postdoctoral researcher at ETH Zurich in Switzerland after having completed his PhD studies in 2013 at Delft University of Technology and VU University Amsterdam in the Netherlands. Evangelos has also been a visiting researcher at EPFL in Switzerland and has industry experience at IBM T.J. Watson Research Center in the USA. Since 2007, he holds a M.Sc. with distinction in Internet Computing from University of Surrey, UK and since 2006 a B.Sc. on Technology Education and Digital Systems from University of Piraeus, Greece. Evangelos has won the Augmented Democracy Prize, the 1st prize at ETH Policy Challenge as well as 4 paper awards and honors. He has published more than 70 peer-reviewed papers in high impact journals and conferences and he is the founder of the EPOS, DIAS, SFINA and Smart Agora projects featured at decentralized-systems.org. He has raised significant funding and has been actively involved in EU projects such as ASSET, SoBigData and FuturICT 2.0. He has supervised several PhD and MSc thesis projects, while he designed courses in the area of data science and multi-agent systems that adopt a novel pedagogical and learning approach. Evangelos’ research interest focus on distributed and intelligent social computing systems with expertise in the inter-disciplinary application domains of Smart Cities and Smart Grids.

This work is supported by the Swiss National Science Foundation (grant no. 170226) for the European FLAG ERA project ‘FuturICT 2.0 - Large scale experiments and simulations for the second generation of FuturICT’ (https://futurict2.eu/).