Project TextMall: Text-Mining Made Simple for All

2 minute read

Published:

The objective of this research is to develop a general-purpose text analytics platform, i.e., Text-Mall, which would enable real-world users to easily explore the power of Text-Mining in a simple and interactive fashion without worrying about the underlying details of Natural Language Processing.

Team TextMall


Overview of Project TextMall

Despite great progress in the field of natural language processing (NLP), computers are still far from being able to accurately understand unrestricted natural language. Indeed, how to develop a completely automated text analytics system that can support many different Text-Mining applications is a very hard problem; as such, Text-Mining technology is not still directly usable by real world users like doctors, teachers, executives etc. Thus, it is essential to involve humans (real users) in the loop in an interactive fashion and facilitate explorative Text-Mining to enable end-users to apply this technology effectively in the real world.

The objective of this research is to develop such a general purpose text analytics platform which would enable real-world users to easily explore the power of Text-Mining in a simple and interactive fashion without worrying about the underlying details of NLP. The PI aims to address this challenge by developing a new general-purpose text analytics platform, i.e., Text-Mall (aka. Text Mining for all), which will provide simple intuitive operators for interactive text-mining tasks which can be easily explained and taught to the general public.

Here is an example use-case of TextMall agent:

Image not Loading


This is the high level architecture of TextMall agent:

Image not Loading


Impact of Project TextMall

Text-Mining is essential for optimizing all kinds of decisions related to people, ranging from making effective and acceptable public policies by governments, to deciding critical medical treatment for patients by doctors, to providing personalized tutoring materials to students by teachers, and/or to effective advertising of products to people on the Internet by companies. Text-Mall, once realized, will enable real world users to conduct such analysis and thus, facilitate more informed decision making process for them. Text-Mall operators will also be combined flexibly to support a wide range of analysis tasks that may require different workflows, thus enabling an application developer to program a text mining application by using Text-Mall as a programming language for text analysis.