الفهرس 
chapter 1Only 14 pages are availabe for public view
 |  | from | 134 |  |  |
| | | from | 134 | | |
Abstract
Data mining has become an increasingly important branch of computer science to deal with the rapid growth of data that have been collected and stored in databases. It is the process of selecting, exploring and modeling large amounts of data. It has resulted in the discovery of useful hidden patterns from massive databases.
Optimization is a mathematical method that relates to find the optimal solution. It becomes a very important methodology appear in scientific life. It can be applied in many different application fields, like telecommunications, data mining, design, combinatorial optimization, power systems and Electronic circuits. There are different types of algorithms that deal with data mining; however all of them can not deal with hidden patterns, so we will use Hybrid between Rough Set and Particle Swarm Optimization (HRSO) to deal with this problem. Particle swarm optimization (PSO) is a relatively new technique; that is based on swarm intelligence; it is inspired by the social behavior of organisms such as fish, bird and ants. Performance optimization experimentally proved to perform well in many problems of optimization use (PSO) algorithm.
Rough set theory (RST) is a mathematical tool to deal with uncertain, imprecise and hidden data. It introduced for data attribute reduction and delete the Redundant features of this data, so it can reduce the complexity of any problem.RST considered an effective approach to reduce the dimensions of the data information system, By deleting attributes from the dataset of knowledge that are not necessary in the data and don’t affect on the result. In the processing of the data, redundant features (attributes, Objects and attribute values, etc) and redundant rows are been deleted based on rough set theory.
Our problem focus on the electronic circuits that is developing every day, this problem is the distribution loads on the grid therefore on users that can cause the cut of electric current because of overloading on the network, and another problem appeared that is the design of circuits is becoming a complex process that needs some simplification. That may be difficult to be done by using traditional way. So we applied HRSO algorithm on the electronic circuit to form it in simple design and optimize its components to avoid the cut of electric current thus, we are working to redistribute the load on the network, therefore we used two algorithms in this thesis; the first one is to reduct the components of the electronic circuits that is called rough set algorithm and the second one is to optimize the components and to find the optimal solution this called particle swarm optimization (PSO).
In this thesis, first we applied the Hybrid Rough Set and Particle Swarm Optimization (HRSO) algorithm that proposed for electronic circuit simplification. The HRSO is applied to simplify circuit by reducing the components of circuit and try to find optimal value of circuit components, and then we make evaluation by the traditional PSO algorithm on the data of electronic circuit then compare the result the proposed algorithm. The result shows that we reach to our goal (7.5V) fast when we use the proposed HRSO algorithm.
We applied rough set by two ways their results show that three resistance {R7, R1, R4} are not necessary because they don’t affect the output of circuit, the results of two ways are similar but rough set program is applied on all data but the manual way applied on a sample of data because it has taken a long time and they have a complex. Then we assume that our goal is to reach to 7.5 V that the optimal solution we will reach to it by three methods; the first one is the manual way that takes a long time and has a complex computations, the second way is PSO algorithm that is applied without RST also takes a long time and is not efficient to approach to our target and the third one is HRSO algorithm that applied RST first then applied PSO the result become more efficient to approach our target. So, our experimental results show that the HRSO algorithm is more efficient than the two ways and it takes less time to obtain the optimal solution.