الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 113 |  |  |
| | | from | 113 | | |
Abstract
The research shows the use of a hybrid stock-picking technique that integrates neural network, fuzzy logic and various quantum techniques to perform on the fluctuations and the movement of the correlations in stock price changes, to increase the efficiency of stock trading while using a suitable model that added advantage for a fund manager with many clients, each with a different portfolio and with variable probability for risk.
Many classical soft computing approaches have successfully applied in the prediction of stock price and showed a good performance. This research investigates the power of Quantum Genetic Algorithm in a neuro-fuzzy system composed of an Adaptive Neuro Fuzzy Inference System (ANFIS) controller used in prediction of stock market, identified using an optimization technique based on a double chains quantum genetic algorithm. In this work the ANFIS controller and the stock market process model inputs are chosen based on a comparative study of many different combinations of past stock prices to determine the stock market process model inputs that return the best stock trend prediction for the next day in terms of the minimum Mean Square Error (MSE). The proposed model are tested with actual financial data and show the weak form of compared approach by demonstrating much improved predictions by finding optimal value for optimization variable in ANFIS using a double chains quantum genetic algorithm. The researcher aimed to predict the gold price in the Forex market, it introduces the use of Quantum Differential Evolution Algorithm in ANFIS controller used in prediction of stock market, specified by an optimization technique based on a double chains quantum differential evolution algorithm, to evaluate the proposed model three performance measurements are used: Root Mean Squared Error (RMSE), percentage error and Mean Tendency Error (MTE). The algorithm was evaluated with actual financial data and proved the weakness of the comparative method by showing much improved predictions by finding the best value for optimization variable in ANFIS using a double chains quantum differential evolution algorithm.
Aiming that the results will show that the hybrid system takes advantage of the synergy among these different techniques to intelligently generate more optimal trading decisions, allowing investors to make better stock trading decision.
Data Mining (DM) is interested in finding hidden relationships in business data to allow the work of the business forecasts for future use. This research can offer an overview of a preliminary technology (data mining). After explaining the meaning of (data mining) and the evaluation requirements, discover the relationship between data warehousing and data mining. And the differences between the tools used here the questionnaire in the database and data mining. The research describes in detail the different stages of the process of exploring data and different models of data mining. Methodologies for the exploration mission for the data were evaluated comparative strengths and weaknesses. I have been using neural network and fuzzy logic framework for action contained in data mining.
So the importance of applying of framework for generating modified fuzzy rules is useful in the database. The process will be carried out to estimate the grade owed to verify the quality and accuracy of the application of neural network technique or methods on tuning parameters of the problem. And this work was also prepared to discover classification rules to discover these rules, developed of new algorithm concepts based on some of the concepts of exploration data, especially with an emphasis on developing of discovered knowledge to work through the dynamic environment.
Time series is considered one of the major classes of data objects; it can easily be collected from financial and scientific applications such as prices of market stocks or daily temperature. Time series is a collection of observations orderly arranged and measured over equal time interval, this fixed rate is the key difference between time series and data stream. Time series has the characteristics of being large in size, high number of dimensions; it is continuously updated, and always treated as whole not individual elements.