You are on page 1of 1
A Computational Model to Evaluate Honesty in Social Internet of Things Upul Jayasinghe Depariment of Computer Science Liverpool John Moores University Liverpool, United Kingdom u.ujayasinghe@2015.ljmu.ac.uk ABSTRACT Trust in Socal Intemet of Things has allowed to open new horizons in collaborative networking, particularly by allowing objects to communicate with their service providers, based on their relationships analogy to human world, However, strengthening trust is a challenging task as it involves identifying, several influential factors in each domain of social-eyber-physical systems in order to build a reliable system. Tn this paper, we address the issue of understanding and evaluating honest that isn important twust metric in trustworthiness evaluation process in social networks, Fitst, we identify and define several ust ateibues, which affect direct to the honesty. Then, a subjective computational model is derived based on experiences of objects and opinions from friendly objects with respect to identified sributes. Based onthe outputs ofthis model final honest lev s predicted using regression analysis. Finally, she effectiveness of our ‘del is tested using simulations CCS Concepts + Human-centered computing-Social network analysis» Computing methodologies~Reasoning about belief and knowledge = Computing methodologies-Modeling ‘methodologies Keywords Social Networks, SloT, Trust Metric, Trust Attributes, Trust Computation, Knowledge, Subjective Models, Regression, 1. INTRODUCTION With the technologicsl advancement in present information and communication infastructates, users and ownets of ther (objets) generate significant amount of social information like followers, Itiends, communities, et. as wel as counterpart of nework level interactions, Based on this, Social Internet of Things (SIOT) concept is formalized, which creates social networks between all peries (humans and objects) and among them like in online social networks (OSN) [4; 1d]. Objects in SIoT autonomously generate Permission to make digital or hard copes ofall or part of his work & perma o elssrom un gant wilt fe provided that copies are bot made or derbi for profit or commercial advange ata copies tear thi tice and the fll cation onthe fist page. Copyrights for componeuts ofthis wrk owned by others than ACM rus. be honored crating with credit is permed. To copy cheroise or republish, to on servers or toredibute ols regues prio speifc permassion ollars bo, Rogen pernissicas fou Peaisonaflerraar. SACT?, Api 47,2017, Makes, Moroso. © 2017 ACM, ISBN 978-1-4503-11869/1 704. $15.0 DOI: htp://dx.doi.org/10.1145/3019612.3019840 Hyun-Woo Lee Hyper-connected Communication Research Laboratory, ETRI Daejeon, Korea hwlee@etri.re.kr Gyu Myoung Lee Department of Computer Science Liverpool John Moores University Liverpool, United Kingdom g.mlee@ljmu.ac.uk feltonships among them inorder to solve common problems in eyber-physical-social systems (CPSS) including service, resource snd network discovery [17], Trowever, heterogensity of the devices, networks and social relationships makes CPSS vulnerable to treats. Te control these kind of situations the concep of trusts introduced [3]. So far, many proposals have been presented on evaluating and managing trust in SloT but yet prototypes lacks the basic explanations on how the information or behavioral data from CSS is collected, processed snd obtain meaningful result in the decision making. process. However, authors in [9] and [18] explain about a trust model and trust computation techniques which are more relevant to our work here, Particularly, inthis paper we focus on evaluating one specific trust attribute (TA) “Honesty” of abject in SIoT which helps to get tan insight about “Knowledge” trust metric (TM) in ou model [18]. We propose subjective model to evaluate honesty in SloT ring many aspects like abject relaionships, spatial and temporal properties of objects and their history of ‘behaviors. The major contributions of this paper are t:() identify tributes, which affect to honesty (i) prevent w numerical model to analyze them, and (li) evaluate the effectiveness of the rnumerical results om real world data set. To the best of out Knowledge, we are the fist 10 propose a subjective method to evaluate honesty in SloT The remainder of the paper is organized as follows, In Section 2, we investigate current contnibutions on ust modeling. ‘management end computation methods. Sections 3 provides abasic ides bout SloT environment and a brief introduction about our past contributions, which provide foundation for this researeh. Based on the definitions, development of numerical model is presented in Section 4 and simulation results based on the numerical model is discussed in Section 5, Finally, Section 6 concludes the paper with «8 summary and future work, 2. RELATED WORK Basic undying issues in tst assessment in lo are the lack of concrete definition of tus, impsct of relationships and context awareness on tus properties and objective of rust management in CPS environment. In this resed, authors in [21] and [16 have provided a comprehensive survey on trast management in JT and presented a strong research model with curent challenges an trust Evaluation, A holistic view of tos in several application domains including TMs and usability in decision making process ie explained in dtl in [65 A preliminary idea of establishing social relationships is firstly introduced in [7] and a more comprehensive description is ‘resented in [3]. On the other hand, [8] and [14] are discussed about trust assessment of social networks based on concepts. like community of interest, fiendship, followers as well as frequency,

You might also like