Stock market prediction is an important topic in ?nancial engineering especially since new techniques and approaches on this matter are gaining value constantly. In this project, we investigate the impact of sentiment expressed through Twitter tweets on stock price prediction. Twitter is the social media platform which provides a free platform for each individual to express their thoughts publicly. Specifically, we fetch the live twitter tweets of the particular company using the API. All the stop words, special characters are extracted from the dataset. The filtered data is used for sentiment analysis using Naïve bayes classifier. Thus, the tweets are classified into positive, negative and neutral tweets. To predict the stock price, the stock dataset is fetched from yahoo finance API. The stock data along with the tweets data are given as input to the machine learning model to obtain the result. XGBoost classifier is used as a model to predict the stock market price. The obtained prediction value is compared with the actual stock market value. The effectiveness of the proposed project on stock price prediction is demonstrated through experiments on several companies like Apple, Amazon, Microsoft using live twitter data and daily stock data. The goal of the project is to use historical stock data in conjunction with sentiment analysis of news headlines and Twitter posts, to predict the future price of a stock of interest. The headlines were obtained by scraping the website, FinViz, while tweets were taken using Tweepy. Both were analyzed using the Vader Sentiment Analyzer.

In recent years, digital marketing has surpassed traditional marketing as the preferred technique of reaching customers. Researchers and academics may utilize it for social media marketing and for predicting client buy intent, among other applications. It can boost customer happiness and sales by facilitating a more personalized shopping session, resulting in higher conversion rates and a competitive advantage for the retailer. Advanced analytics technologies are utilized in conjunction with a dynamic and data-driven framework to expect whether or not a customer will make a purchase from the organization within a certain time frame. To increase income and stay ahead of the competition, one must understand customer buying habits. Several sectors offered rules to explore a consumer's potential based on statistics results. A machine learning algorithm for detecting potential customers for a retail superstore is proposed using an engineering approach.

The act of digital marketing uses a variety of traditional methods such as analyst consensus, earnings per share estimation, or fundamental intrinsic valuation. Also, social media management, automation, content marketing, and community development are some of the most popular uses for digital marketing. Stock price prediction is a challenging task since there are so many factors to take into account, such as economic conditions, political events, and other environmental elements that might influence the stock price. Due to these considerations, determining the dependency of a single factor on future pricing and patterns is challenging. The authors examine Apple's stock data from Yahoo API and use sentiment categorization to predict its future stock movement and to find the impact of “public sentiment” on “market trends.” The main purpose of this chapter is to predict the rise and fall with high accuracy degrees. The authors use an artificial intelligence-based machine learning model to train, evaluate, and improve the performance of digital marketing strategies.

No abstract