Mining frequent phrases from social media

Funding Details
Natural Sciences and Engineering Research Council of Canada
  • Grant type: Engage Grants Program
  • Year: 2015/16
  • Total Funding: $25,000
Principle Investigator(s)

No researchers found.


Project Summary

One of the main aspects of a mobile keyboard application is to predict the most possible intended next word as a user types a sequence of either words or letters. This prediction has to be performed accurately. An accurate prediction suggests words, which a user intends to choose as the next word. In fact, a prediction is accurate if the users choose the next word from the suggested list. Language modeling is the de facto approach for supporting this type of language prediction. In this research project, we are interested in designing a highly accurate language models for a mobile Keyboard platform, based on the state of the art techniques in Natural Language Processing and Information Retrieval. We will build a training corpus by adding new vocabularies, which are not seen in a Standard English corpus to facilitate typing in social networks environment such as Twitter.

Related Projects

Given the pivotal significance of understanding customers' interests through systematic monitoring and analysis of social media platforms for call centers, this proposed research project will address important research challenges related to social me... More ...
It has been reported that 90% of all the data in the world has been generated over the past two years either through user-generated content or by scientific or industrial entities. Analytics software applications, large-scale and data-intensive, can ... More ...
This project focuses on developing techniques within a unique platform for product knowledge base development, management and use. Our work is based on the following assumptions: i) The Web is a rich environment for extracting product information as ... More ...