Trust networks infotech2010

Trust Networks:
Interpersonal, Social, and Sensor
Krishnaprasad Thirunarayan, Pramod Anantharam,
Cory Henson, and Amit Sheth
Kno.e.sis - Ohio Center of Excellence in Knowledge-enabled Computing
Wright State University, Dayton, OH-45435
2/18/2011 Trust Networks: T. K. Prasad 1

Broad Outline
• Real-life Motivational Examples (Why?)
• Trust : Characteristics and Related Concepts (What?)
• Trust Ontology (What?)
– Type, Value, Process, Scope
• Gleaning Trustworthiness (How?)
– Practical Examples of Trust Metrics
• Research Challenges (Why-What-How?)
– Sensor Networks
– Social Networks
– Interpersonal

Real-life Motivational Examples
(Why track trust?)

Interpersonal
• With which neighbor should we leave our
children over the weekend when we are
required to be at the hospital?
• Who should be named as a guardian for our
children in the Will?

Social
• In Email:
– SUBJECT: [TitanPad] Amit Sheth invited you to an
EtherPad document.
– CONTENT: View it here:
http://knoesis.titanpad.com/200
• Issue: Is the request genuine or a trap?

Social
• To click or not to click a http://bit.ly-URL
• To rely or not to rely on a product review
(when only a few reviews are present)?

Sensors
• Weather sensor network-based prediction of a potential
tornado in the vicinity of a city.
• Issue: Should we mobilize emergency response teams
ahead of time?
• Van’s TCS (Traction Control System) indicator light came
on intermittently, while driving.
• Issue: Which was faulty: the indicator light or the
traction control system?
• Van’s Check Engine light came on, while driving.
• Issue: Which was faulty: the indicator light or the
transmission control system ?

Common Issues and Context
• Uncertainty
– About the validity of a claim or assumption
• Need for action
• Critical decision with potential for loss
– Past Experience : Vulnerability Examples
• Irresponsible / selfish guardian => Marred future.
• Illegal invitation / attachment => Loss of private data.
• Malfunctioning sensor => Loss of funds.

Why Track Trust?
• To predict future behavior.
• To incentivize “good” behavior and
discourage “bad” behavior.
• To detect malicious entities.

Trust and Related Concepts
(What is trust?)

Trust Definition : Psychology slant
Trust is the psychological state
comprising a willingness to be
vulnerable in expectation of a
valued result.
2/18/2011 Trust Networks: T. K. Prasad
Ontology of Trust, Huang and Fox, 2006
Josang et al’s Decision Trust
12

Trust Definition : Psychology slant
Trust in a person is a commitment to
an action based on a belief that the
future actions of that person will
lead to good outcome.
Golbeck and Hendler, 2006
13

Trust Definition : Probability slant
Trust (or, symmetrically, distrust)
is a level of subjective probability
with which an agent assesses
that another agent will perform
a particular action, both before
and independently of such an
action being monitored …
Can we Trust Trust?, Diego Gambetta, 2000
Josang et al’s Reliability Trust
14

Trustworthiness Definition :
Psychology Slant
Trustworthiness is a collection of
qualities of an agent that leads them
to be considered as deserving of
trust from others (in one or more
environments, under different
conditions, and to different degrees).
http://www.iarpa.gov/rfi_trust.html
15

Trustworthiness Definition :
Probability slant
Trustworthiness is the
objective probability that the
trustee performs a particular
action on which the interests
of the trustor depend.
Solhaug et al, 2007
16

Trust vs Trustworthiness : My View
Trust Disposition
Depends on
Potentially Quantified Trustworthiness Qualities
+
Context-based Trust Threshold
E.g.*, In the context of trusting strangers, people in
the West will trust for lower levels of trustworthiness
than people in the Gulf.
*Bohnet et al, 5/2010
17

(Community-based) Reputation
• Reputation* is the community or public
estimation of standing for merit,
achievement, reliability, etc.
• Reputation** is the opinion (or a social
evaluation) of a community toward a
person, a group of people, or an
organization on a certain criterion.
• Cf. Brand-value, PageRank, eBay profile, etc.
*dictionary.com
20
**Wikipedia

Trust vs. (Community-based)
Reputation
Reputation can be a basis for trust.
However, they are different notions*.
• I trust you because of your good reputation.
• I trust you despite your bad reputation.
• Do you still trust Toyota brand?
*Josang et al, 2007
21

Trust vs. (Community-based)
Reputation
Trust :: Reputation
::::
Local :: Global
::::
Subjective :: Objective
(Cf. Security refers to resistance to attacks.)

Reputation is Overloaded
Community-based Reputation
vs.
Temporal Reputation-based Process
(Cf. Sustained good behavior over time elicits
temporal reputation-based trust.)

Trust vs. Belief
• Trust is a relationship among agents.
• Belief is a relationship between an
agent and a statement.

Trust Ontology
(What is trust?)
Illustration of Knowledge Representation and Reasoning:
Relating Semantics to Data Structures and Algorithms

Example Trust Network -
Different Trust Links with Local Order on out-links
• Alice trusts Bob for recommending good car
mechanic.
• Bob trusts Dick to be a good car mechanic.
• Charlie does not trust Dick to be a good car
mechanic.
• Alice trusts Bob more than Charlie, for
recommending good car mechanic.
• Alice trusts Charlie more than Bob, for
recommending good baby sitter.
*Thirunarayan et al, IICAI 2009
27

Digression: Illustration of Knowledge
Representation and Reasoning
• Abstract and encode clearly delineated “subarea”
of knowledge in a formal language.
– Trust Networks => node-labeled, edge-labeled
directed graph (DATA STRUCTURES)
• Specify the meaning in terms of how “network
elements” relate to or compose with each other.
– Semantics of Trust, Trust Metrics => using logic or
probabilistic basis, constraints, etc. (SEMANTICS)
• Develop efficient graph-based procedures
– Trust value determination/querying (INFERENCE
ALGORITHMS)

(In recommendations)
(For capacity to act)
(For lack of
capacity to act)

Trust Ontology*
6-tuple representing a trust relationship:
{type, value, scope, process}
Type – Represents the nature of trust relationship.
Value – Quantifies trustworthiness for comparison.
Scope – Represents applicable context for trust.
Process – Represents the method by which the value is
created and maintained.
trustor trustee
*Anantharam et al, NAECON 2010
30

Trust Ontology:
Trust Type, Trust Value, and Trust Scope
 Trust Type*
 Referral Trust – Agent a1 trusts agent a2’s ability to
recommend another agent.
 (Non-)Functional Trust – Agent a1 (dis)trusts agent a2’s
ability to perform an action.
 Cf. ** trust in belief vs. trust in performance
 Trust Value
 E.g., Star rating, numeric rating, or partial ordering.
 Trust Scope*
 E.g., Car Mechanic context.
*Thirunarayan et al, IICAI 2009
** Huang and Fox, 2006
31

Trust Ontology:
Trust Process
 Represents the method by which the value
is computed and maintained.
 Primitive (for functional and referral links)*
 (Temporal) Reputation – based on past behavior.
 Policy – based on explicitly stated constraints.
 Evidence – based on seeking/verifying evidence.
 Provenance – based on lineage information.
 Composite (for admissible paths)**
 Propagation (Chaining and Aggregation)
*Anantharam et al, NAECON 2010
**Thirunarayan et al, IICAI 2009
33

Trust Ontology
34

Bob is a car
aficionado
Alice
Bob
Charlie
Dick
type: referral
process: reputation
scope: car mechanic
value: TAB
type: non-functional
process: reputation
scope: car mechanic
value: 3
Dick is a
certified
mechanic
type: functional
process: policy
scope: car mechanic
value: 10
ASE certified
type: referral
process: reputation
scope: car mechanic
value: TAC
TAB > TAC
Example Trust Network illustrating Ontology Concepts

Unified Illustration of Trust Processes
Scenario : Hiring Web Search Engineer - An R&D Position
Various Trust Processes :
• (Temporal) Reputation-based: Past job
experience
• Policy-based: Scores on screening test
• Provenance-based: Department/University
of graduation
• Evidence-based: Multiple interviews (phone,
on-site, R&D team)

Gleaning Trustworthiness :
Practical Examples
(How to determine trustworthiness?)

Direct Trust : Functional
Reputation-based Process
(Using large number of observations)

Using Large Number of Observations
• Over time (<= Referral + Functional) :
Temporal Reputation-based Process
– Mobile Ad-Hoc Networks
– Sensor Networks
• Quantitative information
(Numeric data)
• Over agents (<= Referral + Functional) :
Community Reputation-based Process
– Product Rating Systems
• Quantitative + Qualitative information
(Numeric + text data)

Desiderata for Trustworthiness
Computation Function
• Initialization Problem : How do we get initial value?
• Update Problem : How do we reflect the observed
behavior in the current value dynamically?
• Trusting Trust* Issue: How do we mirror uncertainty
in our estimates as a function of observations?
• Law of Large Numbers: The average of the results obtained from a
large number of trials should be close to the expected value.
• Efficiency Problem : How do we store and update
values efficiently?
*Ken Thompson’s Turing Award Lecture: “Reflections on Trusting Trust”
40

Beta Probability Density Function(PDF)
x is a probability,
so it ranges from 0-1
If the prior distribution of p is
uniform, then the beta
distribution gives posterior
distribution of p after
observing a-1 occurrences
of event with probability p
and b-1 occurrences of the
complementary event with
probability (1-p).

a= 5
b= 5
a= 1
b= 1
a= 2
b= 2
a= 10
b= 10
a = b, so the pdf’s are symmetric w.r.t 0.5.
Note that the graphs get narrower as (a+b) increases.

Beta-distribution - Applicability
• Dynamic trustworthiness can be
characterized using beta probability
distribution function gleaned from total
number of correct (supportive) r = (a-1)
and total number of erroneous
(opposing) s = (b-1) observations so far.
• Overall trustworthiness (reputation) is its
mean: a/a +b

Why Beta-distribution?
• Intuitively satisfactory, Mathematically precise, and
Computationally tractable
• Initialization Problem : Assumes that all probability values
are equally likely.
• Update Problem : Updates (a, b) by incrementing a for
every correct (supportive) observation and b for every
erroneous (opposing) observation.
• Trusting Trust Issue: The graph peaks around the mean, and
the variance diminishes as the number of observations
increase, if the agent is well-behaved.
• Efficiency Problem: Only two numbers stored/updated.

Direct Trust : Functional
Policy-based Process
(Using Trustworthiness Qualities)

General Approach to Trust Assessment
• Domain dependent qualities for determining
trustworthiness
– Based on Content / Data
– Based on External Cues / Metadata
• Domain independent mapping to trust values
or levels
– Quantification through aggregation and
classification

Example: Wikipedia Articles
• Quality (content-based)
– Appraisal of information provenance
• References to peer-reviewed publication
• Proportion of paragraphs with citation
– Article size
• Credibility (metadata-based)
– Author connectivity
– Edit pattern and development history
• Revision count
• Proportion of reverted edits - (i) normal (ii) due to vandalism
• Mean time between edits
• Mean edit length.
Sai Moturu, 8/2009

(cont’d)
• Quantification of Trustworthiness
– Based on Dispersion Degree Score
(Extent of deviation from mean)
• Evaluation Metric
– Ranking based on trust level (determined from
trustworthiness scores), and compared to gold
standard classification using Normalized
Discounted Cumulative Gain (NDCG)

Example: Websites
• Trustworthiness estimated based on criticality
of data exchanged.
• Email address / Username / password
• Phone number / Home address
• Date of birth
• Social Security Number / Bank Account Number
• Intuition: A piece of data is critical if and only
if it is exchanged with a small number of
highly trusted sites.

Indirect Trust : Referral + Functional
Variety of Trust Metrics
(Using Propagation – Chaining and Fusing over Paths)

Trust Propagation Frameworks
• Chaining, Aggregation, and Overriding
• Trust Management
• Abstract properties of operators
• Reasoning with trust
• Matrix-based trust propagation
• The Beta-Reputation System
• Algebra on opinion = (belief, disbelief, uncertainty)
Guha et al., 2004
Richardson et al, 2003
Josang and Ismail, 2002
63
Massa-Avesani, 2005
Bintzios et al, 2006
Golbeck – Hendler, 2006 Sun et al, 2006
Thirunarayan et al, 2010

Research Challenges
(What-Why-How of trust?)
HARD PROBLEMS

Generic Directions
• Finding online substitutes for traditional cues
to derive measures of trust.
• Creating efficient and secure systems for
managing and deriving trust, in order to
support decision making.
Josang et al, 2007
68

Sensor Networks

Abstract trustworthiness of sensors and
observations to perceptions to obtain actionable
situation awareness!
observe perceive
Web
“real-world”
T
T T
Our Research

Strengthened Trust
Trust

Concrete Application
• Applied Beta-pdf to Mesowest Weather Data
– Used quality flags (OK, CAUTION, SUSPECT)
associated with observations from a sensor
station over time to derive reputation of a sensor
and trustworthiness of a perceptual theory that
explains the observation.
– Perception cycle used data from ~800 stations,
collected for a blizzard during 4/1-6/03.

0
0.2
0.4
0.6
0.8
1
1.2
3/31/2003 0:00 4/1/2003 0:00 4/2/2003 0:00 4/3/2003 0:00 4/4/2003 0:00 4/5/2003 0:00 4/6/2003 0:00 4/7/2003 0:00
Mean Beta
Value
Time
Mean of beta pdf vs. Time (for stnID = SBE)

Research Issues
• Outlier Detection
– Homogeneous Networks
• Statistical Techniques
– Heterogeneous Networks (sensor + social)
• Domain Models
• Distinguishing between abnormal phenomenon
(observation), malfunction (of a sensor), and
compromised behavior (of a sensor)
– Abnormal situations
– Faulty behaviors
– Malicious attacks

Social Networks

Our Research
• Study semantic issues relevant to trust
• Proposed model of trust/trust metrics to
formalize indirect trust

Our Approach
 Trust formalized in terms of partial orders
(with emphasis on relative magnitude)
 Local but realistic semantics
 Distinguishes functional and referral trust
 Distinguishes direct and inferred trust
 Direct trust overrides conflicting inferred trust
 Represents ambiguity explicitly
Thirunarayan et al , 2010

Practical Issues
• Refinement of numeric ratings using
reviews in product rating networks
– Relevance : Separate ratings of vendor or about
extraneous features from ratings of product
• E.g., Issues about Amazon’s policies
• E.g., Publishing under multiple titles (Paul Davies’ “The Goldilock’s
Enigma” vs. “Cosmic Jackpot”)
– Polarity/Degree of support: Check consistency
between rating and review using sentiment
analysis; amplify hidden sentiments
• E.g., rate a phone as 1-star because it is the best 

Research Issues
• Determination of trust / influence from
social networks
–Text analytics on communication
–Analysis of network topology
• E.g., follower relationship, friend relationship, etc.
• Determination of untrustworthy and
anti-social elements in social networks

Research Issues
• Evolving trust ontology
• Introducing trust threshold
– For binary decision to act in spite of vulnerability/risk
• Structuring trust scope
– Class hierarchy
• Structuring trust value
– Or does relative trust suffice?
• Refining trust types
– Or does trust scope suffice?
• Restrictions on trust propagation
– Limited horizon

Research Issues
• Improving Security : Robustness to Attack
– How to exploit different trust processes to detect
and recover from attacks?
• Bad mouthing attack
• Ballot stuffing attack
• Sleeper attack
– Temporal trust discounting proportional to trust value
– Using policy-based process to ward-off attack using
reputation-based process
• Sybil attack
• Newcomer attack

Interpersonal Networks

Research Issues
• Linguistic clues that betray
trustworthiness
• Experiments for gauging interpersonal
trust in real world situations
– *Techniques and tools to detect and amplify
useful signals in Self to more accurately predict
trust and trustworthiness in Others
*IARPA-TRUST program

Research Issues
• Study of cross-cultural differences in
trustworthiness qualities and trust thresholds
to better understand
–Influence
• What aspects improve influence?
–Manipulation
• What aspects flag manipulation?

Conclusion
• Provided simple examples of trust (Why?)
• Explained salient features of trust (What?)
• Showed examples of gleaning trustworthiness
(How?)
• Touched upon research challenges for
gleaning trustworthiness in
• Sensor Networks
• Social Networks
• Interpersonal Networks

Thank You!

Trust networks infotech2010

More Related Content

Similar to Trust networks infotech2010 (20)

Recently uploaded (20)

Trust networks infotech2010