Automatic generation of references

In order to efficiently navigate in topic based documents it is essential to generate navigational references. They serve to point to relevant content areas, thereby creating an information mesh that supports readers with quick access to information. However these relations often are not captured in the first place by authors, leading to an increased time spent on navigating to the desired location within large documents. This presentation is about the automatic generation of such references, thus providing advantages for both authors and readers. To this end techniques to extract key words and different ways of determining similarity are employed to find potential relations between topics. In addition a flexible way to calculate the value of key words is presented that is useful when working with different types of documents. There is an existing and working implementation as a Java library and integration as a plugin into a widely used authoring tool.

Who is presenting?

Ferdinand Pyttel
Bio Coming Soon




Aaron’s biography includes work as a business analyst, consultant, programmer and DITA practitioner. Ferdi’s studies includes information retrieval and work on service oriented architectures.

Return to Agenda