just do a google scholar search on "chinese word segmentation", you will see hundreds of published academic papers on this subject. More recent research efforts seem to focus on machine learning and statisitical-based approaches.
Your demo seems to suggest a linear scan and match-on-the-go algorithm, true?