Senior Economist Julian Chan has developed a framework that uses machine learning to “read” large volumes of text and detect subtle, structural changes embedded in it.
As a first use case, Dr. Chan and co-author Dr. Weifeng Zhong of the American Enterprise Institute developed the Policy Change Index (PCI) for China, an algorithm that can predict China’s potential policy changes. Using the information in over 2 million articles in the government’s official newspaper, the PCI was able to predict many critical shifts in Beijing’s policymaking.
The PCI framework has received significant academic interest and media coverage worldwide. Besides academic conferences, Drs. Chan and Zhong have also presented before government agencies and private companies, including the US Naval Research Laboratory, In-Q-Tel, and State Street Corporation. Their research is also scheduled to be presented at the Strata Data Conference and the World Bank Group later this year.
This framework has the potential to be widely applied to detecting changes in judicial opinions, regulatory codes, and lawmakers’ pronouncements, among other arenas.
