One of the main tasks in the field of Artificial Intelligence (AI) is providing means for understanding of natural language. The understanding is often restricted to analysis of specific general meaning of particular instances of language behavior (sentences, documents, etc.). For example, in subfields of AI such as natural language processing, or sentiment analysis, speaker attitudes and emotions expressed in a sentence are in the focus of analysis. A different kind of task from the field of natural language processing, which we focus on in this research, is predicting trends of future events. From the point of view of understanding natural language it is a challenging task, since it assumes extracting information regarding future unfolding of events from the provided instances of language behavior. To make the task feasible we limited the focus of our research to detecting and extracting sentences which refer to the future as potentially the most useful in predicting future unfolding of events. We assumed that future reference sentences can contain such information as background of events, specific expert knowledge, etc., and thus can be useful in future prediction. We propose automatic extraction method for future reference sentences from news corpora using semantic and morphological information. Furthermore we discuss the experiment in which the sentences extracted with the proposed method are applied to predict future unfolding of a set of specific events.