Mutual Context Based Word Prediction

Abstract

Word prediction systems can reduce the number of keystrokes required to form a message. In our daily life we use lots of messengers online to communicate with friends and others. In our daily life chatting is almost inevitable. In recent years the keyboards that we use have a built in structure for predicting and suggesting our next word. These suggestions are helpful in most of the cases. There already has been lots of works done in this regard and researches are still ongoing. One of the mechanisms of next word prediction is the contextual word prediction. Context is defined as, ”Context is any information that can be used to characterize the situation of an entity. An entity is a person, place, or object that is considered relevant to the interaction between a user and an application, including the user and applications themselves.” Our hypothesis is that word prediction models can be more enhanced if we use mutual context between the users as a paramater in word prediction. We also hypothesize that that mutual context based word prediction has great potential in enhancing word prediction increasing communication rate, but the amount is dependent on the accuracy of detecting the mutual context. We show that in a conversation mutual context based word predition model can do better word prediction than traditional word prediction models.