Identification of bar-codes with inductive learning techniques

Approaching to 2000, computer technology have been rapidly developing to catch the information age. As time is very important some new methods should be developed to get better results within a short time. Bar-code systems which include a range of thin and thick bars and spaces between them is a frequently applied method for processing knowledge and easily recall. For example, in the big markets that sell wide range of products computers seek the price and other information using bar codes and monitor them. Also costumers can monitor all the knowledge. The process is performed within short time because of the simple structure of bar codes. Also with international bar code system all the knowledge represented including the writer, publication date, the name of the book etc. is represented in one code instead of long bibliographic identities. The requirements for processing bar codes are not only labels and scanners. Specially equipped computers, softwares, electronic circuit boards and master films are some other basic elements of a bar code system. Before deciding to install such a system, the financial analysis should be done as the first instalment is usually expensive. In this study, it is demonstrated that bar-code systems are very suitable applications for inductive learning. A number of induction algorithms are available but, in this study, RULES3 induction algorithm is chosen because of its more simple implementation structure and some advantages over others. Finally, it can be easily said that Inductive Learning is an alternative way to process bar codes. The system requires less memory to store and call codes, shorter processing time, and is cheaper.