Our hack converts the short form written in English chat to actual English words. We take care of the context while predicting the sentence. Needless to mention that we do not do it by just maintaining a database. Such a system has an obvious disadvantage of failing on seeing new words. We will also illustrate this using the state-of-the-art available implementations of similar convertors.
For example u r gr8 , it will translate it into "you are great".
More examples after implementation.
1. Any translator from one language to another works on pure and correct english only. So if one person from India is talking to someone who understands only Japanese, then only if the person in india writes correct English, the translator would returning more than just jokes
2. For annotating objects from sentences, all available research/implementations require correct English, Else, they fail miserably, And since there is no doubt that social media can be used to predict the world, our normalizer would be able to help annotate twitter, hence help in getting a lot of information that is presently not fully used.