Abstract:

XML schemas can be used as a method to represent the logical and thematical structure of a text. It is a very useful tool both for teachers and students interested on textual criticism. As a representation, an XML schema is the result of the critic’s own perception about what is the meaning of the text, and how that meaning has taken a concrete linguistic form.

Traditional textual analysis, follows a qualitative approach, where knowledge about the text equals to the critic’s personal view on both literature about the text, and the text itself. This personal view is usually conveyed through a scientific paper, a talk at a conference or the like. However, such a representation, is bound to be challenged by other critic’s visions and eventually modified.

While I respect that way of studying texts in which myself have been rised; what I present here is a (maybe not so) subtle modification of it. Adding the power of quantitative analysis to the equation by means of a topic modeling algorithm, it rises up a strong data-based vision of the textual lexical and sintactical structure.

But the results given by use of topic models are somewhat fuzzy. The visual representations generated by them must be properly filtered and tagged by a human, in order for true knowledge, to be produced. In this sense, an XML schema provides a most exciting bootcamp, in which both teacher and student can improve their abilities to sum up data and extract a single vision from multiple possibilities.

Here I present part of the results of my dissertation research on Information Retrieval (IR) methods applied to Ancient Greek texts. XML coding and topic models revealed themselves to be quite powerful tools when combined. As such I propose an example of real use in a educational context.

Keywords:

Education, XML, TEI, philology, topic modeling.