Predictive Analytics intelligent decision-making framework and testing it through sentiment analysis on Twitter data.

The volume of data in companies and in the private sector is already gigantic today. Mobile devices such as smartphones constantly collect data on all possible environmental conditions. When surfing the Internet, everyone leaves an endless digital trail. The Internet of Things (IoT) promises comprehensive networking of all everyday devices and production tools that surround people. Nevertheless, the modern knowledge society has to face the question of whether we are really actively using all the data or whether useful knowledge has increased as a result. Answering this question is not trivial. It is true that today's opportunities for exploring data, for transforming data into information and thus for gaining knowledge from it, are greater than ever. But it is also true that this new knowledge, which consists of the hidden connections in data, does not appear in our mind's eye on its own. We must explore it to bring it to the surface which is related to recognizing patterns in the world of data. In the last step, these patterns must be correctly interpreted. Predictive analytics (PA) is going exactly in this direction. They are currently one of the most important application areas of big data and are seen as the most actionable form of business intelligence (BI). Predictive analytics can be used for a variety of purposes, from predicting customer behaviour in sales and marketing to determining risk profiles for financing. Another widely known application is credit reporting, used by financial institutions to determine the likelihood that customers will repay future loans on time. It can also be used when working with big data in predicting user behaviour and opinion. In this regard, the purpose of this paper is to develop a predictive analytics-driven decision framework based on machine learning and data mining methods and techniques. To test it, we conducted an experiment for predicting sentiments and emotions in social media posts, as well as discussed topics and extracted keywords.