Constructing semantic network for knowledge representation
Text mining is still budding in the field of medicine. However, with rising volumes of scientific literature it has gained significance and is becoming used extensively in the field of research. Through text mining, researchers are equipped to analyse unstructured data. The structured output attaine...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/54347 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Text mining is still budding in the field of medicine. However, with rising volumes of scientific literature it has gained significance and is becoming used extensively in the field of research. Through text mining, researchers are equipped to analyse unstructured data. The structured output attained from text mining is represented as a semantic network. A semantic network is a relationship diagram, showing entities and the relationship between them. These semantic networks can facilitate the discovery of hidden relations between entities. In this project, biomedical texts would be extracted and biomedical concepts would become the entities for which the network would be formed. These networks can be very vast and hence there is a lot of room for discovery. The ability to visualize this data is also an important factor that facilitates research and any user trying to analyse given data. This report details the research behind the project. The project aims to construct semantic networks for various brain diseases and potentially retrieve hidden relations between them. This project also aims to make the visualized output for every brain disease, there is also the possibility of merging diseases to observe relations between them. This project focuses on identifying and implementing efficient methodologies in the creation of a semantic network. A simple summary of the steps involved in this project would be the following: building a corpus, creating a dictionary, developing algorithms for parsing data and then building algorithms to match this data and finally the aim is to visualize such a network. The report also briefly describes challenges faced and future recommendations to improve the project. |
|---|