The protein structure prediction is one of the major areas of interest and development of bioinformatics, theoretical chemistry, biotechnology and it is highly important in medicine. Protein secondary structure (PSS) plays a major role in 3D structure prediction, in the past two decades, an enormous number of works have been developed for protein secondary structure prediction (PSSP), where the accuracy for that particular research problem are not more than 90%. In this case there is a need to improve high accuracy prediction methods. The emanate methods and strategies are analyzed and generalized probabilistic approach. The main aim of this study is how data mining techniques have applied for predicting the protein secondary structure.