OpenNLP

The world of NLP (Natural Language Processing) is a fascinating area of research and development at the crossroads of many fundamental fields, such as Computer Science, Artificial Intelligence, and Linguistics. With such a rich heritage, it is of no wonder that many digital tools have been dedicated to the task of making computers speak human. OpenNLP is one such application, in essence, a machine learning toolkit for processing all text paragraphs that seem natural to humans, but alien to computers.

The modular structure of the application is highly useful for programmers working in this field, situated on the fringes of science; all-in-all, the utility comes with numerous built-in linguistic or otherwise text analysis-related tools, such as a tokenizer, a sentence detector and a part-of-speech tagger.

OpenNLP features a command prompt setup, in which users manually load each tool necessary for the given task. This allows one to follow a highly sequential logic when parsing texts. Other tasks the application can perform include coreference resolution and chunking or parsing text blocks.

All these components allow users to process basic or advanced tasks, from simple text parsing and sentence recognition operations to complex and holistic maximum entropy and perceptron based machine learning endeavors.

Besides the functions already mentioned, one can also employ the built-in training tools, as these allow users to check the model's performance against specific text test paragraphs.

Several algorithms allow the application to recognize punctuation marks and delineate sentences and name recognition is also supported.

Having so many tools does carry with it the risk of baffling users and one of the great features of this program is the dedicated and highly active development community.

This also doubles as a highly comprehensive help manual, with constantly updated topics and FAQ items, and new users are well-advised to bookmark the development page!