Skip to Content

Diego Mollá Aliod

List of Undergraduate and Postgraduate Projects

Below is a list of possible projects for Honours and Masters students at Macquarie University. Please contact me for further details. Also, if you have a project in mind which is not listed here but that is related to my research interests, contact me and chances are that I will be interested too.

These projects represent my current research interests. If you are pursuing a PhD, some of these projects can be extended to PhD projects.

Natural Language Processing for Medical Text

  1. Classification of Medical Research Abstracts
  2. Determine the Strength of Evidence in Medical Text
  3. Find the Structure of Medical Abstracts
  4. Find the Structure of Full Medical Papers
  5. Cluster Medical Research Papers

Classification of Medical Research Abstracts

Medical research takes many forms and sometimes it is not clear in the abstract what type of study is performed. Was it a controlled randomised trial? or a case study? or a systematic review? In this project you will develop methods to automatically classify medical research papers by applying machine learning methods to their abstracts.

Determine the Strength of Evidence in Medical Text

An important piece of information for a medical doctor is to determine how strong is the evidence presented in the medical literature. In this project you will develop methods to automatically determine the strength of evidence of medical research articles. For this you will implement current written guidelines for the appraisal of medical research.

Find the Structure of Medical Abstracts

Many medical abstracts are clearly structured, but others are not. Given the information provided by the structured abstracts, the goal of this project is to automatically determine the inherent structure in those abstracts that have not been explicitly structured.

Find the Structure of Full Medical Papers

There are many medical papers available in HTML format but it is very difficult to determine what parts of the HTML can be used to determine the sections. In this project you will examine the most common HTML resources and provide methods to remove the irrelevant HTML text.

Cluster Medical Research Papers

Given a list of research papers, the goal of this project is to automatically cluster them into groups and determine what are the key concepts of each group.

Natural Language Processing for Financial Text

CMCRC is the Capital Markets Cooperative Research Centre. This is the initiative of a group of researchers, Universities and industry partners. CMCRC focuses on the development of new and innovative technologies for the capital markets domain.

Summarisation of Financial News in Context

CMCRC has a collection of past market alerts and a list of news items that are presumably related to the alert. The goal of this project is to present summaries of the news in context with the topic of the alert so that a financial expert can easily skim over the news items. Techniques to try here include selecting fragments with shared words with the topic (like Google's snippets) and more ambitious methods that use syntactic and semantic information.

Financial News Classification

CMCRC has a list of news classified according to typical news categories, but these categories differ from those of market tickers. The goal of this project is to re-classify the news using as categories the market tickers. Part of this project involves mapping the original taxonomy of news types to the specific one based on market tickers. Conflicts and ambiguities will need to be resolved.