الفهرس 
AbstractOnly 14 pages are availabe for public view
 |  | from | 103 |  |  |
| | | from | 103 | | |
Abstract
Automatic storage and retrieval of information is among the vital topics in the field of information technology. This mechanism is set to maintain this information in terms of protection. Any damage or lack of such a mechanism must be effective and have a large reliability. In order to achieve this goal and to take advantage of the published research in the field of biological computing (bio computing), a mechanism for storing and retrieving information is suggested to mimic the existing mechanism in the human brain, which merges between the locations of such information and its values through hypertext network. This mechanism relies on taking advantage of the properties of DNA and the proteins (component) and represent it for coding the information as well as coding the indices of information through hypernetwork so that there is a correlation between the content and location. The proposed method is based on the development in the field of parallel computing by converting the input information to a specific set of libraries. The distribution of information entered on these libraries and processing performed on one of these libraries are on all the others. As well as determine the degree of cardinality set numbers by specifying the number that represents and it is possible that this number is variable. The information coding using symbols proteins (A, T, C, G), which represents the DNA strand, as well as the indices of that information, hypernetwork nodes that represent the vertices and the labels that distinguish the entered information from the other sets.
The proposed method has been tested on a set of standard images (128 × 128), using a number of libraries (2, 4, 8, 16 and 24), also was selected variable cardinality set numbers (3, 4 and 5). The changing in all of these properties on the performance are discussed in terms of time spent on storage and retrieval. Also it studied the effect of increasing the volume of information entered on the performance by adding the percentage (10%, 20% and 30%, 40%, 50%, 60%, 70% and 80% and 90% and 100%) in both dimensions of length, width, and these increases affect the time it takes and the number of examples needed to represent information in the hypernetwork. The choosing of the same hypernetwork also discussed for more than one information set entered to store, and its impact in the retrieval accuracy are studied the effect of the complexity of this information because there is a loss of information in the case of non-marking for each group.
The results of the proposed method applied on the concepts of DNA Computing and Graph Theory also showed that the effect of the increase of Cardinality number, and the increase in the Library using Logical NOT Gate is notably slow in the process of data retrieval, the results also showed that by using different bulks of information, using Logic NOT Gate showed a high percentage of reduction, averaged 30%. The results also showed in the proposed method, that when storing the information safely by giving a single code for each stored information, for guaranteeing its secure retrieval with high accuracy (100%). The results showed that by not giving that code, the accuracy would be 0.281%.