The process of extracting sentences from text is known as Sentence Boundary Disambiguation (SBD). While this process may initially appear to be simple, there are many complicating factors that ultimately demand more sophisticated approaches, such as using neural networks.
The end of a sentence is typically marked with a period. However, there are other terminators used, such as question marks and exclamation marks. If these were the only considerations, then the process would be easy. However, even limiting the problem to periods, we find that periods are used in many places including abbreviations, numbers, and ellipses. A sentence might use periods such as Mr. Smith, 2.005, or 3.12.18. Ellipses may be simply three periods back, to back or a Unicode character might be used.
Specialized text such as scientific text may contain unusual uses...
