SeSaMe Seminar by Dr Parma Nand

Title: Text Mining - Enabling Institutions and Countries for the Future
Speaker: Dr Parma Nand
Date & Time: 11th July 2017 (Tue), 1.00pm - 2.00pm
Venue: Seminar Room 3, COM1 (#02-12)

Abstract: Over the years it has been proven beyond doubt that knowledge is the ultimate power. In the current environment of ubiquitous internet there is enormous amount of knowledge available, some of it is proprietary, however a vast majority is open source and is available freely for use by anyone on the internet. Most institutions are critically dependent on Information Communications Technology which means that the knowledge required for competitiveness is available in a digital form. In a lot of cases this is only available as naturally occurring texts. The main obstacle in being able to utilize this knowledge has been our inability to computationally process naturally occurring texts, as most of the information on the internet is written as human generated free text.

Apart from the knowledge embedded in online sites, in the last five years there is has been a social revolution in the form of social media. Social media is principally used to exchange information for non-business related purposes, however it has become so ubiquitous and pervasive that businesses have started using it for commercial purposes.

I will talk about the power of being able to automatically process free texts from social media platforms as well as more generally from other types of free texts. The talk will give a background of the discipline of natural language processing and the inherent difficulties in automatically processing a natural language text. There has been substantial advances in language processing tools, however there is vast room for further development in terms of the machine learning algorithms and new linguistic theories. These tools also need to be able to harness the power of the enormous amounts of knowledge available in language resources such as DBpedia, Freebase and WordNet. I will talk about these aspects then conclude by looking into the future for the internet and knowledge harvesting from it.

Bio: Parma Nand is a Senior Lecturer at Auckland University of technology.

His main research area is Artificial Intelligence, particularly in the Natural Language Engineering. His current interests are based on research projects based around information extraction from free texts as well as generation of free texts.

Some of the current projects are semantic searching from medical records, criminal profiling from police records and question answering/chatbots for specific applications. Prior to these he has worked on social media texts for various purposes such as topic changes and sentiment analysis.

Prior to his academic career Parma has worked as a software engineer for various companies in New Zealand. His main areas of expertise lie in designing software in the retail oil and electronic funds transfer industry.