A Writing Assistent Using Language Models Derived From the Web

In the field of Linguistic there exists many powerful tools for measuring the statistic characteristics of words and sentences. These tools rely on a corpus to which the data is compared. In order to get good and meaningful results from the tools available, a suitable corpus is thus needed. As the corpus is the key that ties the tools together, it is of uttermost importance. For most applications, all though not all, a large corpus is useful. This paper presents a solution to using the largest corpus known to man, the Internet. It will show a prototype program using many different linguistic tools on information gathered by the premiere search engine Google.