Comparative Algorithm for Predicting the Protein Localization Sites with Yeast Dataset

Proteins are an important part of the organism and are involved in almost every process in the cell. Many proteins are biochemical enzymes and important to the process of metabolism. This paper proposes protein localization sites in cells. Comparison of learning algorithms such as Logitboost, Tree Decision, Binary Decision Tree, Bayesian Classifier methods. To use as a guideline to improve the best data analytic model. The efficiency and accuracy of the yeast data set are used. The learning algorithms are used to measure performance and finding the value of the data set. This paper used the principle of data mining in segmentation. In this case, there are five classes and nine attributes, respectively, of 921 records. The analysis system of the data is done by correctly classified instances and incorrectly classified instances. The experimental results show that the Logitboost technique gave the highest accuracy rate at 65.65%.