الفهرس | Only 14 pages are availabe for public view |
Abstract Multiple theories has appeared which help to analyze the data and turn it into information to assist in decision-making, and the most important of these theories of neural networks and theories based on fuzzy logic and genetic algorithms, but these theories were not sufficiently with dealing with incomplete data & has been unable to make a decision from it efficiently. So in recent years Rough set theory [RST] has appeared and have had the ability to deal with incomplete data through the construction of decisions tree that have the ability to deal with incomplete data with some extent of efficiency or deal with data efficiently and reduced it . Decision tree is of a particular importance in the analysis of decision issues, which contains a series of decisions or a series of consecutive nature cases.The theory has been applied to real world problems, including many of the pattern recognition, information extraction, and intelligent systems. The Thesis proposes a new algorithm based on RST called [DTCRSCR], for decision-making,To find out the classification of incomplete information systems and to be used in the identification of persons through the iris of the eye, detect intruders on the network and was also used in the analysis of data. Experimental results of decision trees that have been built to take the decision by the [DTCRSCR] show that they tend to be simpler and more precise structures, also they achieved the highest data classification rather than other methods which were used previously and be of a higher accuracy in the detection. |