Abstract:
Calculating the value of the respondent answers to a research questionnaire is the connected and
repeated work that would require extra time and energy. Besides the dull appearance, potential
errors also occur due to saturation. To help the work, we need a computer method to recognize
each letter to present an answer option (e.g. a, b, c, d and e), save it as a pattern and use as a
comparison when reading the questionnaire answer sheets, so that each respondent answer can be
recognized by the computer. To create a software model for this purpose we use the concept of
Artificial Neural Network (ANN) by using the Back Propagation