What's my native language

📄 Project Abstract

Through statistical, computational and qualitive analysis a corpus of written Native English, and corpora of written interlanguage (where the target language is English), I obtained a set of characteristic features for each strand of language. These characteristics were to be relevant to the task of engineering a piece of software which would act as a native language checking facility in a word processor. This facility, would, after a certain amount of English words have been input by the user of the word processor, be able to guess what the native language of the user is. It would do this by calculating descriptive statistics for the text being input by the user and comparing them against inferred statistics within the templates of characteristic features of strand of language, matching features. I developed a piece of software to statistically analyse the data and produce a histogram of the distribution of a word within smaller chunks of a chunks. I used this is conjunction with the ...