![Electronic Thesis and Dissertation Repository](../../assets/md5images/943042f88f25dd8adfffc3aeec6e330c.png)
Degree
Master of Science
Program
Computer Science
Supervisor
Mercer, Robert E.
Abstract
Abstract
In this thesis, the focus is on the topic of “Extracting Scales of Measurement Automatically from Biomedical Text with Special Emphasis on Comparative and Superlative Scales.” Comparison sentences, when considered as a critical part of scales of measurement, play a highly significant role in the process of gathering information from a large number of biomedical research papers. A comparison sentence is defined as any sentence that contains two or more entities that are being compared. This thesis discusses several different types of comparison sentences such as gradable comparisons and non-gradable comparisons. The main goal is extracting comparison sentences automatically from the full text of biomedical articles. Therefore, the thesis presents a Java program that could be used to analyze biomedical text to identify comparison sentences by matching the sentences in the text to 37 syntactic and semantic features. These features or qualities would be helpful to extract comparative sentences from any biomedical text. Two machine learning techniques are used with the 37 roles to assess the curated dataset. The results of this study are compared with earlier studies.
Recommended Citation
Baker, Sara, "Extracting Scales of Measurement Automatically from Biomedical Text with Special Emphasis on Comparative and Superlative Scales" (2019). Electronic Thesis and Dissertation Repository. 6134.
https://ir.lib.uwo.ca/etd/6134