International Journal of Scientific & Engineering Research, Volume 2, Issue 11, November-2011 1

ISSN 2229-5518

Designing and Knowledge Based Expert System for Handling Business Dynamics

Sunita Bansal, Manuj Darbari

Abstract— This paper focuses on creation of Knowledge base for handling business dynamics using domain transformation by applying back propagation network.

Index Terms— Knowledge base, Domain Transfer Technologies

1 INTRODUCTION

——————————  ——————————
nowledge based news computing emphasis on the use and representation of knowledge about any ap- plication. The powerful capability of Neuro- computing system helps in modeling knowledge using Neuro-Fuzzy approach. This paper deals with theoretical models of computation which has learning capabilities of recurrent network models. It also focuses on various busi- ness patterns where we can use knowledge base one of the tool to analyze. There has been number of applications where we can represent Fuzzy finite automata as a tool to solve business decisions. In context of NN learning, prior learning knowledge is designated to any information con- cerning the task domain or the target function. The know- ledge can have different sources: it can be derived from human expert or accumulated by the system from the
previous experience.
Prior knowledge can be used in Neuro-Fuzzy Approach as tool for Design and training. The methodologies fol- lowed for relating consumer's need with technologies, which are conventionally qualitative, preclude a quantifi- cation of these relations, the sole genuine foundation of an economic and financial analysis on which the entrepre- neur can fully appreciate the resources that he will have to omit in an environment where the joint marketing and technology aspects absolutely must be clearly grasped.

————————————————

 Manuj Darbari is working as Associate Professor with the Department of InfomationTechnology in Babu Banarsi Das National Institute of Technology and Management,Lucknow, India E-mail: manujuma@gmail.com

The solution to this problem is influenced by four forces.
The model expert system should provide information about how the process of creating and modifying re- sources influence their value and when the value has been changed.
The model of expert system should provide information about who was responsible for the resources and when.
The expert system should capture the fundamentals of the user's business and filters out those user requirements that are likely to change.

2. PREVIOUS RESEARCHES

The promises of expert system were historically derived from Nevell and Simon's pioneering work on General Problem Solver & GPS.
Newell and Simon's idea of general problem solver was modified by Riesback and Schank in which they highlight the human reasoning power in an expert system. Rao and Luxhoj defined an integrated intelligent manufacturing system which uses several symbolic reasoning system and numerical computation packages. We will be taking the help of these seasoning systems in developing a framework that can provide proper visualization of Business of Dy- namics.

2.1 DOMAIN TRANSFORMATION IN EXPERT SYSTEM

 Sunita bansal is working as Sr. Lectuer in the Detpartment of Information Technology in in Babu Banarsi Das National Institute of Technology and Management,Lucknow, India, E-mail: sunita_bansal301@rediffmail.com

The output of the transformation process is an economic resource that was of business application want to monitor and conferred one of the outputs is a product, but many conversion process to produce other resources. In this scenario business application are interested in planning, monitoring and controlling the work in progress and in- termediate resources.
http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 2, Issue 11, November-2011 2

ISSN 2229-5518

The value chain model for creating a simple business process is shown in figure 1.

Figure 1 : Expert I/P Business Transformation Model
It is important to observe that information can be lost when transform in the domain resulting in a poor classifier. The or- der to select the best possible combination of Port, Tool and Labour, Network selection criteria is used. The detail model consists of various Export 1..... ExportRV based on ontology based Pattern matching.

To properly decide on various resource quantities each train- ing pattern consisting of an input vector and a desired output response vector a Tree is generated which finally detects the resource requirement at Assembly and Inspection.
Knowledge Base Expert System
y = y (p) | y (t) |y (l)
Step 2 : If y (p) > Required Level switch to y (t) If y (t) > Required level switch to y (e)
Else Return to y (p)
Step 3 : Train the Network for next Iteration.
Step 4 : Generate a Rule Base and upgrade at each situa- tion.
Step 5 : Assign Judging and end the iteration.
The resources y (p) , y (t) and y(l) are viewed as set of keywords by lemmatizing their significance weight age.

2.2 INDEXING TO CREATE KNOWLEDGEBASE

Keywords correspond to a unique multiple value attribute. Let n total number of technical objects like y (p), y(t) and y(l) for making the product P each elementary technical object being represented in scalar form. We get the equation as:
P = {y (p), y (t), y (t)} Where
[y (p1), y (p2)... y (l)]
Fuzzy Inference Rules for Weight Selection
Plant
Back Propogation Learning Frame work
Selection Rule
Map Output
1(y (t1), y (t2), .. y (tn)), (y (l1), y (l2)........ y (ln))
In creation of knowledge we will be taking the manufac- turing advances at each and every process of plant given as:
PKI  {MA1P, MA2P, MA3P,............, MANP}
where PKI = Knowledge Index Bar
Figure. 2 : Workflow dynamics of Expert system
The algorithm which can describe the fuzzy control rela- tion is given as:
PKI can also defined as Technology Vector : is the product view from resources and technologies.
We can develop a grid which represent as : [PKI= R0
MA ] where PKI is the Cartesian product function.
Step : Find the desired transfer function of

http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 2, Issue 11, November-2011 3

ISSN 2229-5518

[10] B. Berard, M. Bidoit, A. Finkel, F. Laroussinie, A. Petit, L. Petrucci and Ph. Shnobelen (2001) : System and software Verification. Model

Checking Techniques and tools. Springer.

Figure 3 : Matrix showing the Cartesian product of Resources and Manufacturing advances to create product during one single iteration.

3. CONCLUSION

The strategy used in development of expert system contains the expertise required to solve specific domain related prob- lems. The major strength lies in the fact that the presence of expert is not needed. The creation of knowledge base is the care of the system. The knowledge - base is not to be confused with database. As the knowledge base is represented through problem solving rules fact, predicate calculus etc. These is ex- tensive use of IF - THEN rule.
Each time new rule is examined, it is checked against the cur- rent status of the problem solution stored in database . Two methods are generally used by rule interpreter to search for answers are forward chaining and backward chaining. The final element that we have used in generating a product is the user interface. It is an interface through which user can enter the initial information in the database.

REFERENCES

[1] Parsali H.R., A.K. Kamrani, " Design and Management of Automa- tion Handbook, " University of Louisville, 1988.

[2] Rogers, "choosing the Right Robot for the Job. The Industrial Robot,

13 (1) , 1986 , pp. 33-34.

[3] Towill, D.R., "A production Engineering Approach to Robot Selec- tion," OMEGA, 12 (3) , 1984 , pp. 261- 272.

[4] A. Amo, J. Montero and V. Cutello (1999) : On the principles of fuzzy classification. Proc. NAFIPS Conference, IEEE Press, Piscatway, NJ, pp. 675-679.

[5] Mavceri, J.G., "The Robot Selection Expert Rose, " Journal of Auto- mated Reasonings, (4), 1985, pp. 357 - 390.

[6] A. Amo, D. Gomez, J. Montero and G. Biging (2001): Relevance and redundancy in fuzzy classification Systems. Mathware and soft com- puting 8 : 203-216.

[7] K. J. Arrow (1951, 1964) : Social Choice and individual values, wiley, New York.

[8] A. Bechara, H. Demasio and A.R. Damasio (2003) : Role of the amygdala in decision making. Annuals of the New York Academy of sciences 925, 356-369.

[9] A. Bechara, D. Travel and H. Damasio (2000) Characterization of the

Decision making defict of patients with a ventromedicals profrontal cortex lesions. Brain 123, 2189-2202.

http://www.ijser.org