Stock market data is a high dimensional time series financial data that poses unique computational challenges. Stock data is variable in terms of time, predicting the future trend of the prices is a challenging task. The factors that influence the predictability of stock data cannot
be judged as the same factors may or may not influence the value of the stock all the time. We propose a data mining approach for the prediction of the movement of stock market. It includes using the genetic algorithm for pre processing and a hybrid clustering approach of Hierarchical
clustering and Fuzzy C-Means for clustering. The genetic algorithm helps in dimensionality reduction and clustering helps to create feature vectors that help in prediction.