Soil is a product of weathering. The factors that affect the naming of soil are quite complex. Based on the three main factors (plastic index, liquid limit water capacity, and clay content), and by using a typical neural network - back-propagation model (Yin Hongfeng, Pattern Recognition and Artificial Intelligence, 1990, 3(1), 1-12) the author is trying to give a quantitative soil classification. As a more satisfying result was obtained, the neural network model is expected to be referred to as an effective technique for soil classification.
