Master's Thesis Thomas Grass

Development of a web application to manage and edit semantically annotated texts

Abstract

Semantically annotated text collections are collections of texts, which are able to be read and unterstand not only by human but also by machines. To make text collection interpretable by machines a text analysis must be done. The first step is to seperate text collections into texts, texts into paragraphs, paragraphs into sentences, sentences into parts of sentences and last but not least the parts into single words. After such a seperation the next step is to attach the parts a few attributes. If we think about the word "live" in the sentence "The live is beautiful", the annotated attributes could be "noun" and "subject".

After single sentences, texts or complete text collections are semantically annotated, it is possible to make statements about existing texts or relations to other texts. The result of the text analysis could be simple "Key performance Indicators (KPI)" like the "number of verbs" or more complex KPIs like the "Flesch Reading Ease" as well as links from parts of the analyzed text (indentified by linking words like "see" or "compare") to other texts or text collections.

Especially in the law there is great interest in the analysis and interpretation of new and existing legal texts. Legal texts, like judgements, laws or similar, should be able to be imported from different sources and analyzed. The results of the text analysis should presented by an userfriendly GUI to the enduser.

The aim of this Master Thesis is the Development of a web application to manage and edit semantically annotated texts. The main focus is the usage in the law.