Automatic Summarization Based on Temporal Semantic Annotations
Summaries are especially important today, where there is a lot of information. This helps to get the most important parts of the text quickly. A good solution would be to automatically create summaries for the texts. In this work, annotated articles from the Estonian TimeML corpus are used to create temporal relation graphs based on the temporal relations of events. These graphs are used to automatically create summaries from the articles. The quality of the automatically created summaries is evaluated using the ROUGE-L method.
Graduation Thesis language
Graduation Thesis type
Bachelor - Computer Science