By analyzing the frequency of certain words within mainstream news media from any country, a machine learning algorithm can produce a quantitative “peace index” that captures the level of peace within that country, according to a new study published this week in the open-access journal PLOS ONE.
The language used in media both reflects a culture’s view of the world and influences how people within the culture think and act, the authors say. “Hate speech” can mobilize violence and destruction, but much less is known about how “peace speech” characterizes peaceful cultures and may help to generate or sustain peace, said study coauthor Peter Coleman, executive director of the Columbia Climate School’s Advanced Consortium on Cooperation, Conflict and Complexity.
In the study, the authors used five “peace indices” previously developed by various researchers to capture levels of peace within 18 countries classified as high peace, intermediate peace or low peace. They then collected 723,574 media articles containing 57,819,434 words originating from these countries, all written by local sources and published online in English.
Using only the high-peace and low-peace countries, the researchers used a machine learning model to identify words whose use in the media was associated with levels of peace.
Overall, lower-peace countries were characterized by the higher prevalence of words related to order, control and fear. These words included “government,” “state,” “law,” “security” and “court.” Higher-peace countries were characterized by an increased prevalence of words related to optimism and fun, such as “time,” “like,” “home,” “believe” and “game.” When the researchers applied the trained machine-learning model to media from the intermediate-peace countries that had not originally been included, the model correctly identified the countries as having intermediate levels of peace.
Some of the countries scoring the highest peace levels were Canada, Australia, Ireland and New Zealand. Those with the lowest included Bangladesh, Kenya, Tanzania and Nigeria. Countries in the middle included Ghana, India, Jamaica and the United States.
“We used machine learning to find the words in local news media that best indicate the level of peace in a country,” write the authors, who were led by Larry Liebovitch of Queens College. “In less peaceful countries, news media focus on government and social control, while in more peaceful countries, its focus is on personal preferences and the activities of everyday life.”
The study “offers the potential to build real-time dashboards to help track changes in peace versus conflict speech in news reporting,” said Coleman.
The authors say that their data may have been biased, in that all sources were in English, which means their model may be more reliable in evaluating countries where English is a more common language for news communication than in other languages. Additionally, they say, the method may include biases already integrated into the preconceived peace indices used in the work. Despite the limitations, they conclude that the data serve as a good starting point to further explore the linguistic differences between lower-peace and high-peace cultures.
The study was coauthored by researchers from University of San Francisco and Vista Consulting. Peter Coleman is also a professor of psychology and education and director of the Morton Deutsch International Center for Cooperation and Conflict Resolution at Columbia’s Teachers College
Adapted from a press release by PLOS ONE.
