Indexing based on topic modeling and MATHML for building Vietnamese technical document retrieval effectively

The grow of data on the Internet has brought to people many information and it also opened some important problem in Information retrieval… Along with it, some search engines have developed for user’s purpose. User can retrieve information by content, keyword or anything what they need. However, data on the Internet is too huge, the results feedback is often millions or hundreds millions for each query. Therefore, with the narrow field, we will meet a difficult to find related information, especially technical information that contain formulas. In this paper, we present a method for building Vietnamese technical text based on topic modeling and MathML for indexing. System has built and tested with over 500 Vietnamese technical text shown that, this system satisfied users’ requires in accuracy and speed. © ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2016.