APPLICATION OF SEQUENTIAL PATTERN MINING FOR STOCK TREND PREDICTION

Abstract

Stock trend prediction is an essential support for investors. Accurate and fast prediction is being applied by researchers using various models. The method of prediction by mining historical data, the candlestick chart is one of the technical analysis tools used by investors to create a stock trading strategy. In particular, the application of data mining to predict stock trends is a new approach. In this paper, we propose a model using data mining techniques to predict stock trends. The predictive model is based on a sequential pattern mining algorithm on a historical data set of stocks. Identifying patterns through similarity is also presented in the paper. Experimental data were collected on https://finance.yahoo.com. The experimental results of the proposed model have better average accuracy than traditional models such as SVM and LSTM.