Docstoc

MANET Security issues at Network Layer

Document Sample
MANET Security issues at Network Layer Powered By Docstoc
					     JOURNAL OF INFORMATION, KNOWLEDGE AND RESEARCH COMPUTER
                      SCIENCE AND APPLICATIONS


                  PIONEERING EXPERT SYSTEMS :
                   INCORPORATION TECHNIQUES
                                  DR. JAYESHKUMAR M. PATEL

Associate Professor, Sanakalchand Patel College Of Engineering, City: Visnagar State:
                                     Gujarat

                                    jayeshpatel_mca@yahoo.com

ABSTRACT-An expert system refers to a computer system which exhibits the human expert's intelligence.
An expert system handle real world problems requiring the expert's involvement, uses a computer model of
expert knowledge and expert reasoning and u comparable with or even superior to a human expert in
performance (accuracy and efficiency}. MYCIN and DENDRAL are two expert systems in history.
Research on the expert systems has slowed down relative to their development since mideighties. This is in
part due to fact that powerful knowledge-based techniques like the rule-based technique have been mature
and the concept of expert systems has merged into many disciplines. On the other hand the neural network
approach, which resurged in the last decade, seems to have opened a new direction for expert system
development, called the integration of neural networks and conventional expert systems. Therefore the main
focus of this paper to explore such integration techniques.

Keywords: Expert System (Es), Knowledge Base(K.B), Inference Engine (I.E.), Neural Networks.

INTRODUCTION:                                          from the inference methods it employ. The terms
The basic building blocks of the expert system are     "expert systems" and "knowledge - based systems
shown in the Fig. 1. An expert system typically        "are sometimes used interchangeably.
consists of a knowledge base, an inference engine, a
user interface and an explanation facility. The        KNOWLEDGE ACQUISITION
knowledge base stores the domain knowledge             The domain expertise that needs to be transferred to
                                                       an expert system is a collection of definitions,
                                                       relations, spec-ialized facts, procedures and
                                                       assumptions. The transfer of the knowledge from
                                                       some knowledge source to a computer system is
                                                       called knowledge acquisition. To acquire
                                                       knowledge from human experts is known as
                                                       knowledge engineering. And to extract the human
                                                       expert's knowledge via interviews or tools is called
                                                       knowledge elicitation. The three models of
                                                       knowledge acquisition defined by Buchanan and
                                                       Shortlists are
                                                       (a) Handcrafting: Code knowledge into program
                                                       directly.
                                                       (b) Knowledge Engineering: Work with an expert
                                                       system to organize his/her knowledge in a suitable
                                                       form for an expert system to use.
                                                       (c) Machine learning: Extract the knowledge from
                                                       training examples. Knowledge acquisition can be
                                                       divided into following 5 stages:
Figure 1 : Basic Elements of an Expert System          1. Identification: Define an appropriate problem and
the inference engine reasons with this knowledge for   determine the characteristics.
solving the problems. Expert system is connected       2. Conceptualization: Find the concepts (objects,
with conventional software such as database            relations, information etc.) to represent the
management system. Power of the expert system          knowledge.
derives from the knowledge it possesses rather than

ISSN: 0975 – 6728| NOV 10 TO OCT 11 | VOLUME – 01, ISSUE - 02                               Page 21
     JOURNAL OF INFORMATION, KNOWLEDGE AND RESEARCH COMPUTER
                      SCIENCE AND APPLICATIONS

3.         Formalization: Choose a knowledge              B, then C." When the condition is satisfied, the
representation method and an inference mechanism.         action is executed.
4. Implementation: Formulate knowledge in the             The advantage of production systems includes
chosen formalism (rules, frames etc.)                     modularity, uniformity and naturalness. Rules have
5. Testing: Verify the knowledge and validate the         been found quite useful for providing the
system.                                                   explanations as to why a question is asked, how a
The whole process is iterative. Throughout the            conclusion is reached, and what strategy is used by
process, the knowledge engineer works with the            an expert system. The disadvantages are the
domain expert closely. Knowledge engineering              inefficiency of the execution if good control knowledge
tools automate the knowledge acquisition. Another         is unavailable and the difficulty of representing the
advantage of using tools is rapid prototyping. Various    algorithmic knowledge.
expert system tools are available now days are ART,       C.      Semantic Nets
KEE, LOOPS, OPS5 and so on. Tools at the high end         A net consists of nodes and link between the nodes.
provide a mixed environment (combination of               Nodes may be the objects, events, concepts. Links
different knowledge representation languages) and         specify the relationship between the nodes.
graphical interfaces and are suitable for large-scale     Inference can be made in semantic nets by
applications. Tools at the lower end are much less        Intersection search, properly inheritance, Graph-based
versatile and for small- scale applications.              matching.
KNOWLEDGE REPRESENTATION                                  The advantage of the semantic net representation
Knowledge representation is a major issue in              lies in the explicit and succinct association between
building expert system. It is concerned with both         objects and concepts. However, without extending
the storage of knowledge in proper data structure and     its formalism, it is difficult to represent
the use of knowledge in intelligent processes.            quantification, disjunction, and implication.
There are 4 levels of knowledge representation:           D.      Frames
(a)       The first level is the implementation level,    A frame is a collection of slots that characterize an
which concerns the possibility of building a              object. Each slot may be filled with a value, a
computer program for the knowledge representation         default, another frame, or procedures. Embedding
language.                                                 the procedures within a
(b)       The 2nd level is the logic level, which
concerns the logic properties of the knowledge
representation language, such as meaning of the
expressions and the soundness of the associated
inference procedures.
(c)       (c) The 3rd level is the epistemological
level, which concerns the knowledge structure
(e.g. semantics) and the inference strategy of the
knowledge representation language.
(d)       (d) The 4ih level is the conceptual language,
which concerns the actual primitives of the
knowledge representation language.
(e)       Various knowledge representation schemes
are:
A.      Logic
In logic 2 commonly applied rules of inference are:
         Modus Ponens : If "A implies B" and "A is       Figure 2: Semantic Nets
true," then "B is true".                                  frame is called procedural attachment. Since both
         Resolution : If "A is false or B is true" and   procedural and declarative representations have pros
"A is true," then "B is true".                            and cons, frames are intended to combine their
The advantage of logic representation includes            advantages.
generality, naturalness, preciseness, flexibility and     The frames representation has several advantages. It
modularity. The major disadvantage lies in the            is natural for representing the structural objects. It is
separation of representation and utilization and the      more efficient than logic. The default reasoning in
inefficiency for the inference.                           frames is decidable, in logic the default reasoning is
B.      Production Rules                                  undecidable.
The knowledge base of such systems consists of the        E.     Connectionists
rules called productions. Each production rule is put     The long-term knowledge of a connectionist network is
in the form of condition - action pair, e.g.," If A and   encoded as a set of weights on the connections


ISSN: 0975 – 6728| NOV 10 TO OCT 11 | VOLUME – 01, ISSUE - 02                                      Page 22
     JOURNAL OF INFORMATION, KNOWLEDGE AND RESEARCH COMPUTER
                      SCIENCE AND APPLICATIONS

between the units. This approach is used in Neural        A.        Completely Overlapped
Networks.                                                 In this architecture, the system is both knowledge-
WHY REASONING UNDER UNCERTAINTY                           based system and a neural network. It has a dual
1.        In many problem domains, it isn't possible to   nature. The system optimizes its performance by
create complete, consistent models of the world.          combining the strengths of the two forms. Depending
Therefore agents (and people) must act in uncertain       on the need, it can be presented to the user as a
worlds (which the real world is).                         traditional expert system or as a neural network.
2.        We want an agent to make rational               One form can be converted to other through
decisions even when there is not enough                   inherent transition mechanisms. Therefore only one
information to prove that an action will work.            form has to be stored.
3.        Some of the reasons for reasoning under
uncertainty:
(a) True uncertainty: E.g., flipping a coin.
(b) Theoretical ignorance: There is no complete
theory which is known about the problem domain.
E.g.,medical diagnosis.
(c) Laziness: The space of relevant factors is very
large, and would require too much work to list the        Figure 3: Completely Overlapped Systems
complete set of antecedents and consequents.
Furthermore, it would be too hard to use the enormous     B.       Partially overlapped
rules that resulted.                                      The system is a hybrid of K.B. system and a NN,
A.        Bayes's Rule                                    exhibiting the features of both. The components
Bayes's Rule is the basis for probabilistic reasoning     share some but not all of their own internal
because given a prior model of the world in the form      variables or data structures. They often
of P(A) and a new piece of evidence B, Bayes's Rule       communicate through computer internal memory
says how the new piece of evidence doorcases              rather than external data files. An expert network
ignorance about the world by defining P(A\B).             augmented with explanation capability is a partially
B.        Dempster-Shafer's Theory                        overlapped
The Dempster-Shafer theory is a mathematical
theory of evidence based on belief functions and
plausible reasoning, which is used to combine
separate pieces of information (evidence) to
calculate the probability of an event. This theory is
a generalization of Bayesian theory of subjective
probability. The D-S theory handles the disjunction
of hypotheses and can provide hierarchical
diagnosis.
C.        Neural Networks                                 Figure 4: Partially Overlapped Systems
Neural netwoks not only capable of deducing the
useful information but also capable of inducing the       C.       Parallel
knowledge from the data. This capability is due to        A.K.B. and NN work in parallel to solve a common
storage of information in a large number of               problem. Both can be stand-alone systems. The 2
connection weights and use of heuristic knowledge         components do not share their own internal variables.
to adjust the weights properly. Therefore NN can          They communicate through input, output devices
take batter decisions than any other probability          such as data files. For example, in a medical
density functions.                                        diagnostic system, NN analyzes the signal and
HYBRID EXPERT SYSTEMS                                     images, and a K.B. system interprets the clinical
Expert networks refer to neural networks used as          symptoms. And the results are combined
experts in a particular domain. A major weakness
of these systems is that they can't justify their
responses as the traditional expert systems do. Some
solutions have been proposed. One of the solutions is
to build a hybrid system combining neural networks
and rule-based techniques.
Five integration techniques have been identified as
follows:




ISSN: 0975 – 6728| NOV 10 TO OCT 11 | VOLUME – 01, ISSUE - 02                                   Page 23
     JOURNAL OF INFORMATION, KNOWLEDGE AND RESEARCH COMPUTER
                      SCIENCE AND APPLICATIONS

Figure 5: Parallel Systems                              approach of neural network and fuzzy logic
D.       Sequential                                     consists of 5 layers:
A K.B. system and NN operate in sequence to solve       1.      The Input Layer: no activation function
a particular problem. Again both are stand-alone        2.      The Input Fuzzy Layer: the membership
systems and do not share the internal variables.        function
The output of one component is passed on to other       3.      The Conjunction Layer: the min function
for further processing, e.g. a NN is used as a          4.      The Output Fuzzy Layer: the max function
preprocessor for filtering the noise and transforming   5.      The Output Layer: F(a) = a/ ∑Xi σji.
the signals to symbols, which are subsequently          The activation level of an input unit is the value of
processed by an expert system.                          certain input variable in the given instance. The
                                                        input value is passed onto the fuzzy set units, which
                                                        then translate the value into a degree of
                                                        membership as the activation level of a fuzzy set
                                                        unit.
                                                        The Conjunction Layer will take the minimum of
                                                        the inputs it receives from the input fuzzy units
                                                        below. The Output Fuzzy Layer will collect the
                                                        information from 1 or more conjunction units.
                                                        There exists variation at this point. An Output
                                                        Fuzzy Layer may take the maximum or sum of the
Figure 6: Sequential Systems                            inputs.
E.       Embedded
In this integration, either a K.B. component is
embedded within a NN or vice-versa. Inherent
information exchange is expected. However, this
architecture differs from the partially overlapped
architectures in that the system's external feature
is determined by the host component only.




Figure 7: Embedded Systems
The Integration techniques can be categorized
according to the nature of coupling:
1.       Fully Coupled: Corresponding to the
completely overlapped architecture.
2.       Tightly Coupled: Including the partially
overlapped and embedded architectures.
3.       Loosely Coupled: The parallel and              Figure 8: A Fuzzy Neural Network Based on Fuzzy
sequential architecture.                                Rules

FUZZY LOGIC AND NEURAL NETWORKS                         The Output Layer generates the final result by
Based on Zadeh's fuzzy set theory, fuzzy logic          integrating the information from output fuzzy set
views each predicate as a fuzzy set. Fuzzy neural       units. How the output unit calculates its activation
networks are well known for their ability to handle     level depends on the defuzzification scheme.
the fuzzy nature of inference involving symbols. In     The fuzzy neural network can learn by adjusting
fuzzy logic, a linguistic variable like "size" can      the weights. Different training strategies include:
have several linguistic values like "small",            •       Backpropagation
"medium", or "large". Each linguistic value is          •       Reinforcement
viewed as a fuzzy set associated with a                 •       Statistical methods such as random weight
membership function. The degree of membership           change combined with annealing.
can be interpreted as the degree of possibility,        CONCLUSION
which evades the requirement of satisfying the          Es consists of K.B. which separate from inference
probability axioms. The architecture of hybrid          and control components, contains expert knowledge

ISSN: 0975 – 6728| NOV 10 TO OCT 11 | VOLUME – 01, ISSUE - 02                                 Page 24
     JOURNAL OF INFORMATION, KNOWLEDGE AND RESEARCH COMPUTER
                      SCIENCE AND APPLICATIONS

coded in some form as the production rules, frames
etc. In this paper we studied the important knowledge
acquisition & knowledge representation techniques in
an attempt to study how ES& NN can be related.
Five integration techniques have been discussed.
Out of these completely overlapped architecture is
the most advanced form of integration.
Combination of fuzzy logic & NN has resulted in
extremely powerful computation model known as
FUZZY NEURAL NETWORK. Representing a
linguistic value as a fuzzy set has enabled the
system to deal with many expert problems.
REFERENCES:
[1].      Jacek M. Zurada, (2003), "Artificial Neural
Networks," JAICO publication.
[2].       Limin Fu, (2003), "Neural Networks in
Computer Intelligence," McGraw-IIill.
[3].       Dan W. Patterson, (2006), "Introduction to
Artificial Intelligence and Expert Systems," Prentice-
Hall, India.
[4].       Daniel E. O. Leary, (June 1990), "Expert
System Security," IEEE Trans.
[5].      Michael K. Wick and James R. Slagle,
(1989), "An Explanation Facility for Today's
Expert Systems," IEEE Trans.
[6].      Detlef Nauck, Frank Klawonn & kruse,
(1993), "Combining Neural Network & Fuzzy
Controller", Germany.
[7].       Neural Networks at Work: Computer
Applications, IEEE, 1993.




ISSN: 0975 – 6728| NOV 10 TO OCT 11 | VOLUME – 01, ISSUE - 02   Page 25

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:13
posted:1/7/2012
language:
pages:5
jianghongl jianghongl http://
About