ABSTRACT
This chapter contains several advanced applications of natural language processing that often solve a problem completely, or are used in conjunction with other applications to form a larger product. Information extraction and text-transforming applications are discussed and implemented in this chapter.
Information extraction refers to identifying key pieces of information from text. There are various types of information extraction tasks. This chapter contains a variety of implementations for keyphrase extraction and named-entity recognition, including basic implementations, open-source pre-trained models, pre-tuned Transformer models, custom-training models, and fine-tuning Transformer models.
Certain applications output text that is a transformed version of the input text. Examples include document summarization and language detection and translation. In this chapter, different types of text summarization, such as extractive and abstractive, are discussed and implemented using different tools in Python. Service providers, open-source tools, and pre-tuned Transformer models are discussed for the language detection and translation task.
