Isabella Kroon

Media Analysis: Representation of the Fukushima Water Incident

July 2023

Japan's recently announced plans to release treated wastewater from its Fukushima Nuclear Plant back to the ocean has created controversy globally. In this project, I produced preliminary findings on how Western, Chinese, and Japanese news sources differed in their representations of the incident.

After collecting news articles in Lexis-Nexis, I created a program in Python to scrape each of the articles to collect the article's title, media source, date, text, and keywords and to write this data into a CSV file. The media sources were sorted into four categories: Chinese media, Japanese media, Western media, and other media. Next, I created vizualizations of the data to answer the following questions:


  1. Per month, how frequently were media sources covering the Fukushima water incident?

    4 scatterplot graphs with trendlines, frequency of articles per month separated by media source

  2. How frequently were media sources discussing the environment or politics when covering the Fukushima water incident?

    4 trendline graphs showing the frequency of environmental and political keywords by media source

  3. What proportion of articles covering the Fukushima water incident discussed the environment or politics?

    4 trendline graphs showing the proportion of articles with environmental or political keywords per month, separated by media source

  4. Using unsupervised machine learning, what topics were frequently mentioned together?

    topic models, 10 groups of words

Reflection:
This project allowed me to further systematize the data scraping program I developed for my first Research Assistant project so that it could be applied to any set of RTF collection. In addition, I got more involved with the media analysis aspect of my professor's work as we investigated how a politically-charged event was covered differently around the world.

×