Sentiment analysis is a task of authors’ opinion extraction towards objects mentioned in text. The constant and rapid growth of information makes manual analysis literally impossible. Initial approaches originating from microblogging shot text analysis. Such texts mostly tend to convey a single opinion towards the product or service, and hence could be treated as-it-is in analysis. However, once it comes to in a way more larger documents, the provided analysis is expected to be granular. In this talk we cover the advances of machine-learning approaches in sentiment analysis of large mass-media documents. We provide both evolution of the task over time including a survey of task-oriented models starting from the conventional linear classification approaches to the applications findings of the recently announced ChatGPT model.