Newswise — Two University of Arkansas at Little Rock researchers have used a novel approach to quantitatively portray legislative change in Ukraine to provide a view into the larger political dynamics of the country. 

Zachary Stine, a Ph.D. student in computer and information sciences, and his mentor, Dr. Nitin Agarwal, UA Little Rock Jerry L. Maulden-Entergy endowed chair and professor of information science, used topic modeling to produce a quantitative picture of legislative change in Ukraine over 12 years.

“This research gives us a quantitative picture that fits what we believe to be a true representation of Ukrainian politics in real time,” Stine said. “If you are a political scientist, you probably aren’t going to read 17,000 draft laws to understand the political situation. This research gives you a window into the political landscape.”

The goal of the research is to quantitatively characterize a political system as an ongoing, unfolding process. The research can be used to explore ideological pathways through a political space, contextualize the voting behaviors of politicians, and trace the evolution of a political system.

“We selected Ukraine as a case study because of the U.S.’s strategic and geopolitical interests in Ukraine as it is a vital NATO member, its proximity to Russia, Russia’s meddling in Ukraine’s political system, Russia’s annexation of Crimea, Russia’s aggression in eastern Ukraine, and increasing socio-politico-cultural influence of Russia in Ukraine,” Agarwal said.

Stine and Agarwal computationally analyzed all draft legislation, more than 17,000 draft laws, produced by the Ukrainian parliament, the Verkhovna Rada, between 2006 and 2018. Many politically important events occurred during this time period in Ukraine, including the 2014 ousting of Viktor Yanukovych as president. The time period encompasses the fifth, sixth, seventh, and the majority of the eight convocations, which represent the tenure of a newly elected parliament.

“The parliament of Ukraine consists of an ever-changing array of political factions in which membership is fluid,” Stine said. “They aren’t like the U.S. where we have two main political parties. In Ukraine, people may become a politician as part of one political party, but they vote with other parties. This creates some interesting political dynamics.”

The researchers used the topic modeling algorithm, latent Dirichlet allocation, to identify word-usage patterns and represent Ukrainian draft laws as a distribution of topics.

“As the textual artifacts of a complex political process, Ukrainian draft laws encode the paths explored through a political space on the part of the Verkhova Rada,” Agarwal said. “By condensing each draft law into a distribution of inferred topics, we can measure how surprising a given law is relative to some number of preceding laws using the notion of novelty.”

One of the study’s findings was that Ukraine’s greatest legislative changes largely deal with how the country manages relationships with foreign countries. Stine and Agarwal found that the Committee of European Integration and the Committee of Foreign Affairs produce draft bills with the highest novelty value on average.

“This is notable in light of the 2013 protests which would lead up to the 2014 revolution and eventual ouster of president Viktor Yanukovych,” Stine said. “These protests were initially motivated by the decisions to break association talks with the European Union, widely seen as a capitulation to Russian interests. The high average novelty of these committees suggests that they have been drivers of legislative innovation and change across these convocations.”

The study was presented at the International Conference on Social Computing, Behavioral-Culturing Modeling, and Prediction and Behavior Representation in Modeling and Simulation in July.

In future research, Stine and Agarwal plan to analyze additional information from legislation approved by Ukraine’s parliament, such as voting records, bill sponsorship, supporting legislative committees, etc. It is hoped that the research one day might be able to identify connections between politicians and political allies and predict how politicians are likely to vote as well as when major political shifts are about to occur.

This research is funded in part by the U.S. Office of Naval Research (N00014-17-1-2675) and the Jerry L. Maulden-Entergy Endowment at UA Little Rock.