izpis_h1_title_alt

Obvladovanje zaupanja v e-okoljih : doktorska disertacija
Zupančič, Eva (Author), Trček, Denis (Mentor) More about this mentor... This link opens in a new window

URLURL - Presentation file, Visit http://eprints.fri.uni-lj.si/2856/ This link opens in a new window

Abstract
Zaupanje je bistvenega pomena za učinkovito sodelovanje, kar pomeni, da medsebojna izbira (poslovnih) partnerjev in posledično sprejemanje odločitev temelji na tem, koliko ti zaupajo drug drugemu. Način vrednotenja zaupanja v e-okolju se razlikuje od tistega v fizičnem svetu, saj je spletno okolje v primerjavi s fizičnim svetom precej omejeno glede razpoložljivosti in prisotnosti indikatorjev zaupanja. V e-okoljih zaupanje vrednotimo z uporabo sistemov za obvladovanje zaupanja in ugleda, ki temeljijo na različnih matematičnih modelih zaupanja in ugleda. Predpogoj za učinkovit sistem za obvladovanje zaupanja in ugleda je, da implementira razumno in smiselno formalizacijo zaupanja in ugleda v e-okoljih. Doktorska disertacija temelji na modelu kvalitativne dinamike ocenjevanja (angl. Qualitative Assessment Dynamics, QAD), ki v definicijo procesa dojemanja in vrednotenja zaupanja vpelje človeške dejavnike. S tem predstavlja preudaren in prepričljiv model zaupanja in ugleda. Na podlagi računalniških simulacij smo model QAD ovrednotili in analizirali. Načrtovali smo različne scenarije, v katerih imajo agenti v e-okolju vloge iskalcev in ponudnikov storitev, pri čemer nekateri ponudniki storitev priskrbijo storitve z nezadovoljivo kakovostjo. Izvedli smo dva niza simulacijskih eksperimentov. Prvi niz prikaže učinek uvedbe preprostega sistema za obvladovanje zaupanja in ugleda v e-okolje ter prikaže osnovne lastnosti vrednotenja zaupanja, kot ga definira model QAD. Model QAD definira t. i. operatorje QAD, ki modelirajo vedenje agentov in postopek vrednotenja zaupanja. Z drugim sklopom poskusov demonstriramo dodatne lastnosti in učinke operatorjev QAD. Analizirali smo vedenje agentov, kot ga definirajo operatorji QAD, in s primerjavo odločitev, ki so jih sprejeli agenti s pripisanimi različnimi operatorji, pojasnili lastnosti operatorjev QAD. Ena od izrazitih težav v sistemih za obvladovanje zaupanja in ugleda je prisotnost neresničnih ocen zaupanja, ki jih v sistem posredujejo zlonamerni agenti. Za rešitev navedenega problema smo predlagali model zaupanja QADE (angl. Qualitative Assessment Dynamics Extended), ki osnovni model zaupanja QAD razširja z elementi, kot so zasebni vektor ocen zaupanja, zgodovinske matrike zaupanja, zgodovinski zasebni vektorji ocen, splošna miselnost agenta, operatorji zaupanja QADE, agent napadalec in funkcija podobnosti sim. S tem model zaupanja in ugleda QADE predpostavlja možnost obstoja neresničnih ocen zaupanja. Hkrati model QADE upošteva človeško naravo zaupanja in predpostavlja, da različne ocene zaupanja lahko pomenijo neresnične (zlagane) ocene, lahko pa izkazujejo razlike v odnosih do zaupanja različnih agentov. Na podlagi modela zaupanja QADE razvijemo metodo za obvladovanje zlaganih ocen zaupanja. Metoda je sestavljena iz dveh delov. Najprej agent ocenjevalec poišče ostale agente v danem e-okolju s podobnim odnosom do zaupanja. Podobnost med agenti se izračuna s primerjavo ocen zaupanja, ki sta jih podala primerjana agenta, in primerjavo t. i. splošne miselnosti primerjanih agentov. Agent ocenjevalec primerja lastne ocene zaupanja z ocenami, ki so jih podali podobni agenti, in s t. i. črno piko oceni tiste, ki so bistveno različne. Ocene zaupanja, ki jih s črno piko označi določeno število agentov, veljajo za nepoštene in so izključene iz nadaljnjega postopka izračuna vrednosti zaupanja. Učinkovitost predlagane metode smo ovrednotili na podlagi večagentnih simulacij. Izvedli smo simulacije v simulacijskih skupnostih z različnim številom agentov napadalcev in napadenih agentov kot tudi z različnimi vrstami napadalcev – skupinskimi in samostojnimi. Rezultati so pokazali, da je naša predlagana metoda v povprečju boljša za 28 % do 58 % v primerjavi z reprezentativno eksogeno metodo, ki jo je predlagal Teacy s soavtorji, in z reprezentativno endogeno metodo, ki so jo predlagali Whitby in soavtorji. Ocene zaupanja, ki jih posredujejo agenti v e-okolju, so subjektivne, saj je zaupanje odvisno tudi od agentovega nagnjenja k zaupanju. Izračun ugleda agentov z združevanjem posredovanih ocen zaupanja je lahko vprašljiv, če dejavnik subjektivnosti pri vrednotenju zaupanja ni ustrezno upoštevan. Da bi vsi uporabniki enako razumeli vrednosti ugleda agentov, je treba posredovane ocene zaupanja ustrezno prilagoditi vsakemu agentu posebej. V zadnjem delu doktorske disertacije obravnavamo ustrezno združevanje in prilagajanje posredovanih ocen zaupanja. Kot rešitev predlagamo razširitve in modifikacije modela zaupanja QAD ter razvijemo metodo HOMRA (angl. Human-Oriented Method for Ratings Adaptation), ki izračuna posameznemu agentu prilagojene vrednosti ugleda drugih agentov z izločitvijo dejavnika subjektivnosti. Z uporabo metode za prilagoditev ocen zaupanja HOMRA imajo vsi agenti primerljive možnosti pri izbiri agentov v nadaljnjih transakcijah, ne glede na njihovo nagnjenje k zaupanju. Predlagana metoda HOMRA z uporabo neparametričnih statističnih testov poišče agente s podobnimi nagnjenji k zaupanju. V naslednjem koraku izračuna posameznemu agentu prilagojeno vrednost ugleda drugega (ocenjevanega) agenta v e-okolju, tako da združi ocene zaupanja do ocenjevanega agenta, ki so jih posredovali agenti s podobnimi nagnjenji k zaupanju. Lastnosti agentovega nagnjenja k zaupanju so izpeljane iz preteklih ocen zaupanja, s katerimi je dani agent ocenil druge agente. Metoda HOMRA od agentov ne zahteva razkritja procesov vrednotenja zaupanja kot tudi ne kriterijev, prepričanj, meril, motivacijskih ali drugih faktorjev, ki vplivajo na proces evalvacije zaupanja. Uspešnost naše metode smo ocenili na podlagi obsežnih simulacij z različnim številom agentov v e-okolju, različno porazdelitvijo njihovih tipov osebnosti in različnim odstotkom razpoložljivih ocen zaupanja. Rezultati so pokazali bistvene izboljšave naše metode v primerjavi z drugimi metodami za prilagoditev ocen zaupanja. V povprečju je metoda HOMRA 50 % učinkovitejša od metode, ki sta jo predlagala Abdul-Rahman in Hailes, in 73 % učinkovitejša od metode, ki so jo predlagali Hasan in soavtorji, upoštevajoč konfiguracije z različnim številom agentov v e-okolju, konfiguracije z različnimi porazdelitvami osebnostnih tipov agentov kot tudi konfiguracije z različnimi odstotki razpoložljivih ocen zaupanja. Ključne besede: zaupanje, ugled, modeliranje zaupanja in ugleda, sistemi za obvladovanje zaupanja in ugleda, zlagane ocene, subjektivnost, človeški dejavniki, e-okolje

Language:Slovenian
Keywords:zaupanje, ugled, modeliranje zaupanja in ugleda, sistemi za obvladovanje zaupanja in ugleda, zlagane ocene, subjektivnost, človeški dejavniki, e-okolje, računalništvo, disertacije
Work type:Doctoral dissertation (mb31)
Tipology:2.08 - Doctoral Dissertation
Organization:FRI - Faculty of computer and information science
Year:2014
Publisher:[E. Zupančič]
Number of pages:XIII, 188 str.
UDC:004.7.056(043.3)
COBISS.SI-ID:1536160451 Link is opened in a new window
Views:511
Downloads:128
Metadata:XML RDF-CHPDL DC-XML DC-RDF
 
Average score:(0 votes)
Your score:Voting is allowed only to logged in users.
:
Share:AddThis
AddThis uses cookies that require your consent. Edit consent...

Secondary language

Language:English
Title:Managing trust in e-environments
Abstract:
Trust is essential to effective collaboration, meaning that (business) partners choose each other and make decisions based on how much they trust one another. The way how to assess trust in e-environments is different from those in physical world, as there are limited indicators available in online environment. In e-environments, trust is evaluated with the usage of trust and reputation management systems (TRMS) that are based on different mathematical trust and reputation models. The underlying mathematical trust and reputation model should be reasonable in order to implement effective TRMS. The thesis is based on the Qualitative Assessment Dynamics (QAD) trust model, which considers human factors involved in trust reasoning and as such presents sound mathematical trust and reputation model. At first, we evaluated the QAD trust model based on simulations. We proposed different scenarios with agents that played service requester and service provider roles, where some of the service providers provide unsatisfying services. Further, we performed two set of experiments. The first set of experiments demonstrates the effect of introducing trust and reputation management system into environment and demonstrates the basic properties of trust evaluation defined within the QAD trust model. The QAD trust model defines different QAD operators that model an agent’s behavior and process of trust evaluation. The second set of experiments shows the properties and the effects of the QAD operators. We analyzed the behavior of the agents modeled with different QAD operators and explained the QAD operators properties based on comparison of the decisions they made. One of the prominent problems within trust and reputation management systems is the presence of unfair ratings, which has not been sufficiently addressed so far. To address the problem of unfair trust ratings we present the QADE trust model, which extends the QAD trust model with elements such as private trust vector, historical trust matrices, historical private trust vectors, agents’ general mindset, QADE operators, attacker agent and similarity function. As such, the QADE trust and reputation model assumes and models the existence of unfairly reported trust assessments. It considers human-centric nature of trust, where differently reported trust values do not necessarily mean false value propagation but can also imply differences in agents’ trust attitudes. Based on the QADE trust model, we provide the method to identify and filter out the presumably unfair trust values. The method is two-part. Firstly, a trust evaluator finds other agents in society that are similar to him. Similarity between agents takes into account pairwise similarity of trust values and similarity of agents’ general mindsets. Secondly, the trust evaluator black-marks reported trust values if they differ with theirs. Trust values that are black-marked by certain amount of agents are considered to be unfair and excluded from trust computation. We compared the effectiveness of methods to decrease the effect of unfair ratings based on simulations. We made the simulations in environments with varying number of attackers and targeted agents, as well as different kind of attackers – individuals and collaborative attackers. The results showed significant improvements of our proposed method with average improvements from 28% to 58% compared to the other most representative filtering methods by Teacy and Whitby. Trust ratings shared by agents in e-environments are subjective as trust evaluation depends on evaluator’s personal disposition to trust. As such, aggregation of shared trust ratings to compute an agent’s reputation may be questionable without proper consideration of rating subjectivity. So that all agents understand the ratings in the same way, the reported trust ratings should be adjusted to each agent individually. In the last part of the thesis, we address the problem of proper trust rating analysis and aggregation. We propose an extension of the QAD trust model, named QADES. We propose a novel Human-Oriented Method for Ratings Adaptation (HOMRA) that derives adjusted reputations compliant with the behavioral patterns of the evaluators. With HOMRA method, all participants have comparable opportunities to choose trustworthy agents in future transactions, regardless of their trust dispositions. The proposed HOMRA method finds the agents with similar trust dispositions, taking advantages of non-parametric statistical methods. After that, it computes the personalized reputation scores of other agents with the aggregation of trust values shared by agents with similar trust dispositions. The method derives the characteristics of agents’ trust dispositions implicitly from their past ratings and does not request them to disclose any part of their trust evaluation process, such as motivating criteria for trust assessments, underlying beliefs, or criteria preferences. We evaluated the performance of our method with extensive simulations with varying number of agents, different distributions of agents’ personality types, as well as simulations with different number of available trust ratings. The results showed significant improvements of our HOMRA method with average improvement of 50% over the Abdul-Rahman and 73% over the Hasan method, when comparing the efficiency of the methods depending on the number of agents in certain e-environment, the efficiency of the methods depending on the distributions of agents with different personality types and the efficiency of the methods depending on the availability of trust ratings.

Keywords:trust, reputation, trust and reputation modeling, trust and reputation management system, false ratings, subjectivity, human factors, e-environment, doctoral dissertations, theses

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Comments

Leave comment

You have to log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back