In this thesis, we propose a design of informative haptic patterns which convey symbolic meanings and its transferring system framework. Previous works like Vibetones or Braille are not very suitable to deliver information because they make people feel difficult to connect a meaning like a word or action with a pattern. In this thesis, we suggest the method that we can make a symbolic pattern with extracting shape similarity of a symbolic graphical icon. Also we propose the system framework which can display it in touch, and then evaluate the patterns with user tests. As a result, the haptic patterns were recognizable and the framework works well for them. Also, in the learning period, people take less time to memorize with the symbolic haptic patterns. When we compare random haptic patterns and the symbolic ones which made with shape similarity, two groups of users show significant difference (p=0.00656$\lt$0.05). We concluded that those symbolic haptic patterns, which we call HIcon (Haptic Icon), are effective to transfer meanings and easy to memorize. The end of this work we describe an example of HIcon application.