Thesis Topic Details
Topic ID: |
2929 | |
Title: |
Intelligent Comparison and Relational Mapping of Textual Information | |
Supervisor: |
Aleksandar Ignjatovic | |
Research Area: |
Algorithms | |
| Associated Staff | ||
|---|---|---|
Assessor: |
Robert Lang | |
| Topic Details | ||
Status: |
Active | |
Type: |
R & D | |
Programs: |
CS CE BIOM BINF SE | |
Group Suitable: |
Yes | |
Industrial: |
Yes | |
Pre-requisites: |
-- | |
Description: |
In the financial domain, there are many announcements from various sources such as the financial press, web, analyst reports and official releases from exchanges. When an announcement is first made by the source, it is picked up by multiple news agents such as newswire services and financial analysts who then interprets these for subsequent release into general circulation. There is potential here for a Chinese Whispers problem where the meaning of the information could be distorted as it is circulated from one source to the next. There is also the potential to combine different viewpoints from each source to gain a more comprehensive view of the announcement. The project will attempt to match these seemingly disparate announcements and provide metrics that help in visualising the degree of closeness and relationship between the news articles. This can be potentially harder than it sounds as synonyms and frequencies of words need to be taken into account, etc. Some techniques from plagiarism detection could be used to detect similarities between the texts. However, existing plagiarism detection techniques do not take context into account, which is where we need to be if we want to form relationships out of these articles from different sources. |
|
Comments: |
This is an industrial topic for students supported by CMCRC only | |
| Past Student Reports | ||
| No Reports Available. Contact the supervisor for more information.
Check out all available reports in the CSE Thesis Report Library. NOTE: only current CSE students can login to view and select reports to download. |
||