New Universal Property Across Literature and Image Analysis

Hsing-Yi Lai^1*, Li-Ming Wang¹, Tzay-Ming Hong¹

¹Department of Physics, National Tsing Hua University, Hsinchu 30013, Taiwan

* Presenter:Hsing-Yi Lai, email:harrylai88818@gmail.com

Statistical analysis has been used to quantify properties in linguistics such as Zipf's law and Heap's law that are universal across different languages. In this work we report a property that is not only as universal as Zipf’s law, but also shared by linguistics, paintings, and pictures. By plotting the first-level units (syllables/pixels in languages/images) against the second-level units (words/principal components), we found a structure composing of multiple layers that can be mapped to a master curve. We propose to add this property in the future as a check for the credibility of the segmentation method in use.

Keywords: Linguistics, Complex System, Principal Component Analysis