Despite being one of the most widely used techniques of financial management, stocks have drawn increasing numbers of investors during recent years. A substantial degree of risk is involved in buying stocks. The foremost concern for investors is how to minimise risks and maximise returns. One of the most common issues in the stock market is predicting a company's stock value. Volatility in individual profits and the health of the economy are negatively impacted by fluctuations in stock prices. One of the most widely held beliefs among humans is that investing in stock markets, which are supposed to produce excellent outcomes, is one of the finest ways to generate money. Volatility in stock market prices can occur for a variety of causes. It fosters an environment of uncertainty, which discourages constructive investment. Stock markets are notorious for their volatility. Those who are directly or indirectly involved in stock markets should be aware of it. It is necessary to create an intelligent system that can make forecasts based on a variety of indications such as fundamental, statistical, and technical trends. However, no single good predictive model has ever been able to consistently outperform market patterns. Traditionally, predictions for time series data have been made based on previous data and market trends, as well as historical correlation data and projections. Above all, there is no system that calculates projections based on a user's choice of investment type and risk tolerance. The main focus of this research work is on predicting stock market price changes. Instead of looking at daily changes, this research examines the price trend over specific time intervals by identifying turning points. To determine the increasing trend of price change, deep neural network model is used for accurate predictions. In this research work, an Efficient Time Series Stock Market Predictions using Time Interval Triggered Flag Attribute Model (ETSSMP-TITFA) using deep learning is proposed that predicts the lower bound and upper bound of stock market price predictions of multiple companies. The proposed model is contrasted with the traditional models and the results represent that the proposed model performance is better.