Recently, Internet applications have been integrated in 3C hand-held intelligent products such as smart phones, vehicle navigation devices and driving recorders that help improve our living quality. Among these products, digital contents which displayed by a small screen plays an important role in displaying information. The research objective is to explore an optimum representation of character contents for hand-held intelligent products having a better viewing and operating interface. In this research, the character contents for study included typeface, size, distance between two words, distance between two lines and character and background color matching. The research first collected and analyzed existing character contents. Three Chinese and English typefaces were chosen. A threshold value was determined for each character attribute. Based on the defined character attributes, the research developed a user interface for experimental simulation. The experiment was implemented with 48 tested subjects and their experimental data were forwarded to the simulation process of back-propagation neural network for determining an optimum display of character characteristics. It is expected that the results will be helpful for developing better digital content interfaces for hand-held intelligent products.