Predicting stock values is a fascinating and difficult area of study. The power economy is used as a benchmark for evaluating the economic development of nations. The stock market is often regarded as a prestigious industry nowadays due to the potential for high returns with little risk. The stock market, with its vast and ever-changing information sources, is a fertile ground for data mining and economic study. To aid investors, managers, decision-makers, and end-users in making sound financial choices, we used a non-linear regression technique and the k-nearest neighbor algorithm to forecast future stock prices for a company's stock data. This method trains the module using the daily open, close, high, and low values and trading volumes of a stock. Then, an initial stock price is obtained from the user and used as a test variable for the component. The stock's expected closing price will be provided by the module. Using a visualization graph created between the actual and forecasted closing prices of the stock, the discrepancies between the two may be understood. The kNN method was shown to be reliable and have a low error rate, therefore the findings made sense and were logical. Furthermore, the prediction results were near to and practically parallel to real stock prices when compared to the actual stock prices data.