Indulgence of GPS Technique in AI & SE
Pavan Kumar P V R

Abstract:

General Problem Solving (GPS) Technique is a procedure with which we can solve any type of problem. It has a vital importance in the fields like Artificial Intelligence , Software Engineering, DBMS, and even in the fields like Mathematics, General Science also.

Introduction:

A problem is an issue which needs some domain knowledge , a few strategies and techniques along with some intelligence grouped together as a component which executes it as a process to get an output that is expected. As in case of Information Technology terminology a problem can be termed as a “Project”, in case of Mathematical & Applied Sciences we call as a problem or an issue that is to be solved by some theorem proving techniques or by performing some experiments.

In general a problem can be stated as a “Query” wherein we speak for a good solution. In order to solve a problem we may opt for any feasible technique that may suit the current environment .For all sorts of problems a technique called “Problem Solving Technique Procedure” can be implemented to get a feasible optimal solution. This General Problem Solving Technique is a Five level process which can be stated as follows:

• Identify the problem

• Analyze the problem precisely

• Identify the alternative solution strategies

• Choose one among the best alternative solution strategy

• Apply the chosen strategy to get an optimal solution

Whenever we apply all these steps in a sequential order we get the derived optimal solution that we expected. In general this can be visualized as follows :

As and when the GPS Technique concept has evolved it has its indulgence with many more branches like Artificial Intelligence, Software Engineering, Intelligent Systems, DBMS etc. The concepts of AI like AI Tasks, AI Problem Solving Strategies & Techniques are being supported by the GPS Technique at its backend. GPS in AI :
“The ability to reason, to trigger new thoughts, to perceive and learn” what we call as Intelligence. Basically, to develop intelligent systems (which possesses Intelligence) which are poor in learning, reasoning can be defined as AI. It is the intelligence endowed with quickness of mind, faculty of reasoning well informed and communicative.

In general AI can be defined as “Developing computer programs to solve complex problems by applications  of processes that are analogous to human reasoning processes”. AI is concerned with qualitative rather than quantitative problem solving, with reasoning rather than calculations, with organizing large and varied amounts of knowledge rather than implementing a single well defined algorithms.

As a whole AI can be defined as a developing the techniques to understand, learn, reason with which we try to solve problems. With the involvement of GPS into AI it has been marked as three different concepts like

Problem Space
State Space
Production System

With which we can identify, analyze and also solve the problem so that we can get an optimal solution.

A Problem Space is a space which identifies all the specific domain requirements of the problem. A State Space is a search space which identifies all the possible states where it identifies the Initial State i.e. where to start the problem and a Goal State i.e. where to end the problem. Finally the Production System which is a governing body which governs the total transaction of the problem by framing a set of rules.

For instance if we take a simple number puzzle problem with a  3X3 grid we can frame the following:

As 3X3 puzzle is concerned it has 9 square slides where 8 of which are filled with digits 1 to 8 and 9^th is left as BLANK for the problem transition, which we consider as a problem space. For the concerned problem there exists more than one initial state and more than one final state. But we can define one initial and one final state depending on which we can frame the production rules. As production rules are concerned we can frame 4 Permanent Rules and 9 Temporary Rules. The rules thus formed are as follows:

Permanent Rules:

• BLANK moves towards UP/TOP                   
• BLANK moves towards DOWN/BOTTOM
• BLANK moves towards LEFT
• BLANK mover towards RIGHT

Temporary Rules:
The ordered pair (x,y) indicates the rows and columns of the grid.
 X indicates the Row position, Y indicates the Column position.

• (x,y) / x=1 and y=1 gives an option that BLANK moves towards LEFT or towards DOWN.

• (x,y) / x=1 and y=2 gives an option that BLANK moves towards LEFT or towards RIGHT or towards DOWN.

• (x,y) / x=1 and y=3 gives an option that BLANK moves towards RIGHT or towards DOWN.

• (x,y) / x=2 and y=1 gives an option that BLANK moves towards UP or towards DOWN or towards RIGHT.

• (x,y) / x=2 and y=2 gives an option that BLANK moves towards LEFT or towards RIGHT or towards UP or towards DOWN.

• (x,y) / x=2 and y=3 gives an option that BLANK moves towards LEFT or towards UP or towards DOWN.

• (x,y) / x=3 and y=1 gives an option that BLANK moves towards UP or towards RIGHT.

• (x,y) / x=3 and y=2 gives an option that BLANK moves towards LEFT or towards RIGHT or towards UP.

• (x,y) / x=3 and y=3 gives an option that BLANK moves towards UP or towards LEFT.

While framing the solution of  8 Puzzle problem  we have synchronized the concepts of GPS with the concepts of AI. Technically both the concepts are correct but terminology is concerned both has a varied difference. With this the GPS technique has a vital importance in AI  problem solving techniques. In some cases it also implements some strategies like Heuristic Strategy and Control Strategy which is again a subset of the Production System.

In general with the solution strategy with GPS we don’t seek for the terminological method but with the concepts of AI we seek for that. As we have a saying that “Languages are different but meanings are same “ similarly terminology’s are different but methodology is same.

GPS in SE :

Software Engineering is the technological and managerial discipline concerned with systematic production and maintenance of Software products that are developed and modified on time and within cost estimates. It involves a practical application , scientific knowledge to the design and constructions of computer programs and its associated documents.

In the Computer Science terminology the problem that we encounter can be termed as project and we do the project with the involvement of sciences like Software Engineering , OOSD, SAD, UML techniques. Even though without these techniques the same project can be solved with the usage of GPS. If we do a project with the concepts of SE we do that with the involvement of LIFE CYCLE Model called a Water Fall Model where it is a FIVE Phased Model consists of the following phases:

Analysis
Design
Implementation
Testing
Maintenance

Which can be synchronized with all the five phases of GPS as follows :
Technically speaking identifying the problem, analyzing the problem and choosing the alternative strategies together can be termed as ANALYSIS of SE. Further phases like Choosing the best strategy and implementing it to get an optimal solution can be logically termed as the phases like IMPLEMENTATION and TESTING of SE.

Thus we can synchronize the phase of GPS with all the prototype models of Software Engineering as well as with the concepts of Artificial Intelligence.

Pavan Kumar P V R
Lecturer
Dept.Of Computers, Aurora’s P.G College
Chikkadapally, Hyderabad.20
pavan_peri@yahoo.com, pavanp@aurora.ac.in