Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Look up keyword
Like this
4Activity
0 of .
Results for:
No results containing your search query
P. 1
Unification Of Randomized Anomaly In Deception Detection Using Fuzzy Logic Under Uncertainty

Unification Of Randomized Anomaly In Deception Detection Using Fuzzy Logic Under Uncertainty

Ratings: (0)|Views: 80|Likes:
In the recent era of computer electronic communication we are currently facing the critical impact of Deception which plays its vital role in the mode of affecting efficient information sharing system. Identifying Deception in any mode of communication is a tedious process without using the proper tool for detecting those vulnerabilities. This paper deals with the efficient tools of Deception detection in which combined application implementation is our main focus rather than with its individuality. We propose a research model which comprises Fuzzy logic, Uncertainty and Randomization. This paper deals with an experiment which implements the scenario of mixture application with its revealed results. We also discuss the combined approach rather than with its individual performance.
In the recent era of computer electronic communication we are currently facing the critical impact of Deception which plays its vital role in the mode of affecting efficient information sharing system. Identifying Deception in any mode of communication is a tedious process without using the proper tool for detecting those vulnerabilities. This paper deals with the efficient tools of Deception detection in which combined application implementation is our main focus rather than with its individuality. We propose a research model which comprises Fuzzy logic, Uncertainty and Randomization. This paper deals with an experiment which implements the scenario of mixture application with its revealed results. We also discuss the combined approach rather than with its individual performance.

More info:

Categories:Types, Research
Published by: White Globe Publications (IJORCS) on Mar 02, 2012
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

03/06/2013

pdf

text

original

 
International Journal of Research in Computer ScienceeISSN 2249-8265 Volume 2 Issue 2 (2011) pp. 7-14© White Globe Publicationswww.ijorcs.orgwww.ijorcs.org 
UNIFICATION OF RANDOMIZED ANOMALY IN DECEPTIONDETECTION USING FUZZY LOGIC UNDER UNCERTAINTY
S.Rajkumar
1
, V.Narayani
2
, Dr.S.P.Victor
3
 
1
 Research Scholar/Bharathiar University, Asst.Prof & Head / CSE, NIET,Coimbatore, India.
2
 Research Scholar, Dept. Of Computer Science, St.Xavier’s College, Tirunelveli, India.
3
 Associate Professor & Head, Dept. Of Computer Science, St.Xavier’s College, Tirunelveli, India.
 Abstract:
In the recent era of computer electroniccommunication we are currently facing the criticalimpact of Deception which plays its vital role in themode of affecting efficient information sharing system. Identifying Deception in any mode of communicationis a tedious process without using the proper tool for detecting those vulnerabilities. This paper deals withthe efficient tools of Deception detection in whichcombined application implementation is our main focus rather than with its individuality. We propose aresearch model which comprises Fuzzy logic,Uncertainty and Randomization. This paper deals withan experiment which implements the scenario of mixture application with its revealed results. We alsodiscuss the combined approach rather than with itsindividual performance.
 Keywords:
Deception, Detection, Uncertainty, Fuzzylogic, Randomness
I.
 
INTRODUCTIONDetection of Deception is useful for managers,employers, and for anyone to use in everydaysituations where telling the truth from a lie can helpprevent you from being a victim of fraud/scams andother deceptions [1].
 A.
 
 Identifying the Deception
Deception detection between relational partners isextremely difficult, unless a partner tells a blatant orobvious lie or contradicts something the other partnerknows to be true [5].
 B.
 
Fuzzy logic
Fuzzy logic is the part of artificial intelligence ormachine learning which interprets a human’s actions.Computers can interpret only true or false values but ahuman being can reason the degree of truth or degreeof falseness. Fuzzy models interpret the human actionsand are also called intelligent systems [7].Fuzzification is the process of changing a real scalarvalue into a fuzzy value. This is achieved with thedifferent types of fuzzifiers. Fuzzification of a real-valued variable is done with intuition, experience andanalysis of the set of rules and conditions associatedwith the input data variables. There is no fixed set of procedures for the fuzzification [6].
C.
 
Uncertainty
Uncertainty must be taken in a sense radicallydistinct from the familiar notion of risk, from which ithas never been properly separated. Although the termsare used in various ways among the general public,many specialists in decision theory, statistics and otherquantitative fields have defined uncertainty, risk, andtheir measurement as follows:1.
 
Uncertainty: A state of having limited knowledgewhere it is impossible to exactly describe existingstate or future outcome, more than one possibleoutcome.2.
 
Measurement of Uncertainty: A set of possiblestates or outcomes where probabilities are assignedto each possible state or outcome.3.
 
Risk: A state of uncertainty where some possibleoutcomes have an undesired effect or significantloss.4.
 
Measurement of Risk: A set of measureduncertainties where some possible outcomes arelosses, and the magnitudes of those losses variables[3].
 D.
 
 Randomness:
The Dictionary of Oxford defines 'random' as"Having no definite aim or purpose; not sent or guidedin a particular direction; made, done, occurring, etc.,without method or conscious choice; haphazard." Thisconcept of randomness suggests a non-order or non-coherence in a sequence of steps or symbols, such thatthere is no intelligible pattern or combination [8].II.
 
PROPOSED RESEARCH MODELThe following figures show the basic and itsexpanded form for the proposed model.
 
 
Figure1: Basic Proposed ModelFigure2: Expanded Proposed Model
III.
 
RESEARCH METHODOLOGY
 A.
 
Fuzzified Anomalies for Our Proposed Research Model:
The interception of fuzzified anomaly in the field of Recruiters selection process can be analyzed as,i.
 
Specify the range of conditions
0
 

(
)=
(
)
1
 Candidate Answer at the time‘t’ holds the membershipfunction.ii.
 
Classification and categorization
Table I: Membership value assignments
Factor-X Membership
value μ
t
(x)Fully knowledged* 0.900 to 1.000Maximized knowledge 0.800 to 0.899Desired knowledge 0.700 to 0.799Sufficient knowledge 0.600 to 0.699Average knowledge 0.500 to 0.599Partial knowledge 0.400 to 0.499Show-off knowledge 0.300 to 0.399Minimized knowledge 0.200 to 0.299Poor knowledge 0.100 to 0.199Null knowledge* 0.000 to 0.099
* Null and fully knowledge of values 0.000 & 1.000are subject to constraints of Ideal machine.iii.
 
Probing the assumptionsIt is a critical thought of identifying the associationsbased on assumptions towards a competitor by thecorresponding recruiter.For example
Figure3: Association Rules sample
Recruiter SelectionAssumption Deceiver=>Association of the following
 
cues identification (verbal and non verbal)
 
Test mode –self explanation
 
Critical questions
 
Concentration on each counter output
 
Usage of Ranking / comparisoniv.
 
Operational rulesIf (More Quantified Data) ThenIf (Gestural Deception) ThenIf (Verbal DD) Then
 
Features of JavaBasePlatform IndependentPolymorphismEncapsulationInheritanceClasses and ObjectsAbstractnessAssociationCorrespondentApplication of Base FeaturesActual DatumPredictabilityExpectationEntropyRandomizationUncertaintyProposed Datum
Fuzzy Membership Value
ProbablenessAllocationAssumptionClassificationCondition
 
Randomized FuzzifiedEnviornment
Legitimacy
 
UniformityProbableAllocationOperation RulesAssumptionFuzzy-AnomalyCategorization
 
AuthorityNeutralityCondition
 
InterviewerIntervieweeSystematicUncertaintyRandomUncertainty
Level 1Level 2Level 3
UncertainFuzzifiedEnviornment
Information Sharing SystemTest for uncertaintyImplementing Randomization techniqueCombining Uncertainty &Fuzzy Logic
 
Applying Fuzzy LogicCombining Uncertainty, Fuzzy Logic and Uncertainty
 
Combining Randomization& Fuzzy LogicCombiningUncertainty &Randomization
 
10 S.Rajkumar, V.Narayani, Dr.S.P.Victor 
If (Non-verbal/modal DD) ThenIf (Contradictory Results) ThenDeception Detection= truev.
 
Allocation of Boolean sets
 
(
)=
  �
(
)
(
)
 
=1
 
=1
+
  �
(
)
(
)
 
=1
 
=1
/2
 N = Number of testing components/ Questions
α
i
= Assumption for an candidate with an initial setting
of α
1
(x) =1 as a deceiver
β
 j
= Non verbal communication
γ
= verbal communication0 <= Alloc
(x) = μ
t
(x) <= 1Where Alloc(x) =1 represents deceiver and Alloc(x) =0 represents non deceiver.vi.
 
Statistical probabilityDeceivers most probably use the recurrencestrategic tokens during their responses. Let usconsider the collection of sentences CR(s) consistingof a sequence of N words such as (r1, r2, …, rN), thenthe probability for the occurrence of CR(s) can becomputed as
(
(
)=
=1(
/
−+1
,….,
−1
)
 where
(
/
−+1
,….,
−1
)
= frequency
(
−+1
,….,
)
frequency
(
−+1
,….,
−1
)
 
 B.
 
 Randomness-Entropy for Our Proposed Research Model:
Shannon denoted the entropy H of a discreterandom variable X with possible values {x1, ..., xn} as,
(
 
)=
(
(
 
))
 Here E is the expected value, and I isthe Information content of X. I(X) is itself a randomvariable. If p denotes the probability massfunction of X then the entropy can explicitly be writtenas.
(
)=
(
)
(
)
=1
=
(
)log
1
(
)
=1
=
(
)log
(
)
=1
 where b is the base of the logarithm used. Commonvalues of b are 2,Euler's number e,and 10, and theunit of entropy isbitfor b = 2,natfor b = e, anddit(or digit) for b = 10.[3]In the case of pi = 0 for some i, the value of thecorresponding summand 0 logb 0 is taken to be 0,which is consistent with thelimit: 
lim
→0
log
=0
 
The proof of this limit can be quickly obtainedapplyingLHospital’s rule. 
(
)=
(
)log
(
)
=1
 
C.
 
 Randomness-Entropy for Our Proposed Research Model:ii) Las vegas Algorithm
Las Vegas algorithm is a randomized algorithm thatalways gives correct results; that is, it always producesthe correct result or it informs about the failure. Theusual definition of a Las Vegas algorithm includes therestriction that the expected run time always be finite,when the expectation is carried out over the space of random information, or entropy, used in the algorithm.The complexity class of decision problems that haveLas Vegas algorithms with expected polynomialruntime is ZPP.(Zero-error Probabilistic PolynomialTime) It turns out that

=

−1
 class. RP-Randomized Polynomial complexity classand its inverse as CO-(RP) or RP-1 which is intimatelyconnected with the way Las Vegas algorithms aresometimes constructed. Namely the class RP israndomized polynomial time consists of all decisionproblems for which a randomized polynomial-timealgorithm exists that always answers correctly whenthe correct answer is "no", but is allowed to be wrongwith a certain probability bounded away from onewhen the answer is "yes". Thus Las vegas plays itsvital role in decision making.
 D.
 
 Random Uncertainty Evaluation for Our Proposed  Research Model
The uncertainty has a probabilistic basis andreflects incomplete knowledge of the quantity. Allmeasurements are subject to uncertainty and ameasured value is only complete if it is accompaniedby a statement of the associated uncertainty .Theoutput quantity denoted by Z is often related to inputquantities denoted by X1, X2,…,XN in which the truevalues of X1, X2,…,XN are unknown. Then theuncertainty measurement function Z(x) = f(X1, X2, …,XN) Consider estimates X1, X2, …, XN respectivelytowards X1, X2,…, XN based on certificates, reports,references, alarms and assumptions. Each Xi ~prob. DistributionX
1
_X
2
__ _
 
Z(x) = X
1
+X
2
 

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->