Risk Assessment of Stock Market Analysis using Time Series Analysis

Stock market forecasting and risk assessment heavily influence investors' financial decisions. Many depend on news announcements to judge the purchase or sale of risky stocks. Due to the complexity and ambiguity of natural languages, however, reliable modeling and risk management of stock market patterns derived from news releases is challenging. In contrast to past work in this field, which generally uses bag-of-words to extract tens of thousands of characteristics to construct a prediction model, this study employs a novel approach that extracts tens of thousands of features directly from the text itself. We present a Time Series Forecasting based method for financial/stock market prediction with risk assessment using a histology dataset. In particular, Time Series Forecasting is performed at the pre-processing stage to extract time period-related characteristics from financial news. Using the collected features, a time seriesbased metaheuristic CNN (MCNN) model is used to construct a prediction with risk assessment. Using a model based on MCNN, we hope to get better outcomes than previous methods in terms of speed and precision while minimizing risk. The performance is attributable to the time analysis performed during the pre-processing step since it decreases the feature dimensions substantially. Using the suggested method, we aim to enhance the accuracy of stock price forecasts based on a selection of data sets.