Conformance of Chinese text to Zipf's law

摘要

An investigation was carried out to determine whether Chinese text material conforms to Zipf's law. The information reservoir for this particular investigation contains 2,022,604 Chinese ideograms. It is shown that single Chinese characters do not conform to Zipf's law; however compound words are found to conform well. In addition, examining the regression analysis for compound words implies a good degree of conformity.