Fujitsu Analyzes Japanese Election Data with Foundry from Palantir Technologies
TOKYO, Oct. 18, 2021 — Fujitsu Limited conducted a joint data analysis project with Nikkei Inc. (Head Office: Chiyoda-ku, Tokyo; President: Tsuyoshi Hasebe), leveraging data analysis tools including Foundry from Palantir Technologies in order to provide new insights into the Japanese electoral process.
Drawing on the knowledge of Nikkei, the joint project aims to provide readers and voters with in-depth coverage that combines data from previous election results with region-specific data, including employment rates, demographic data, etc. as well as information about political candidates running for office in parliamentary elections. The insights obtained from these analyses will be published in the Nikkei starting October 17, 2021.
By gaining new insights from data, Fujitsu will further strengthen its efforts to drive “Digital Shifts” towards data-driven management models with the potential to innovate customers’ businesses, industries, and society.
In contrast to many countries in the world, where news outlets provide extensive, data-centric analysis for major political events like presidential elections, Japanese media coverage often focuses more narrowly on the results of the elections rather than offering a broader analysis of the context and data informing citizens’ voting behavior and the final results of the election.
Data like voting rates, election results, and election related expenses in Japan are often difficult to analyze as electoral districts are not identical with administrative districts and electoral districts are frequently redistricted. This makes it challenging to offer coverage on the “bigger picture,” as data is hard to match with region-specific information like the demography and employment rate of administrative districts and information about candidates in the lower house elections.
To help address these challenges, Fujitsu and Political News Group of the Policy News Unit of Nikkei, Japan’s leading financial newspaper, jointly utilized a broad range of information including election related data held by Nikkei and data open to the public to analyze previous election results and voting behaviors in order to provide readers and voters with a new perspective through data-driven political coverage of elections.
About the Joint Project
With this project, Fujitsu’s data scientists analyzed past election results by combining publicly available election related data and census data, chosen based on the knowledge of Nikkei. For these analyses Fujitsu leveraged the data analysis tool “Foundry” from Palantir Technologies Inc., which empowers organizations to rapidly process large amounts of data. By using this tool, Nikkei’s political reporters can draw new insights from publicly available data.
This approach offers Nikkei’s political reporters a new tool to create contents that offer readers and voters greater nuance and new insights into topics like the effects of regional characteristics on voting behavior or the effects of economic conditions on electoral outcomes.
The project further aims to give journalists access to objective, empirical data to reinforce conclusions that have conventionally been based on their own experience and intuition. By adding such new, additional value to the media coverage of elections, the joint project aims to contribute to even greater interest and participation in politics from readers and voters in the future.
1. Publication media:
The Nikkei morning edition, The Nikkei Online Edition, etc.
Analysis: August to October 2021
Publication: Starting October 17, 2021
Insights related to the following two topics will be published in the upcoming articles
Theme 1: Regional characteristics and their effect on election trends and results
The focus of this analysis will be the question of how regional characteristics of electoral districts influence election results.
This analysis aims to clarify the characteristics of voting behavior of residents in the relevant electoral districts by analyzing attribute data like age, gender, occupation and education in relation to data about population decrease, economic indicators, and past election results in the respective districts.
Theme 2: Characteristics of council members and their effects on election trends
Based on the characteristics of election winners, analysis will be conducted to help identify possible factors (including age, position and electoral funds) that enabled candidates to win elections. This analysis will also focus on factors like reputation, support groups, and “financial muscles” of candidates.
Fujitsu will draw on the experience gained through this project to offer customers further solutions to leverage complex valuable data in order to contribute to “Digital Shifts” towards a data-driven, innovative management paradigm with the potential to transform not only customers’ businesses but also industries and society as a whole.
Fujitsu is the leading Japanese information and communication technology (ICT) company offering a full range of technology products, solutions and services. Approximately 126,000 Fujitsu people support customers in more than 100 countries. We use our experience and the power of ICT to shape the future of society with our customers. Fujitsu Limited reported consolidated revenues of 3.6 trillion yen (US$34 billion) for the fiscal year ended March 31, 2021. For more information, please see www.fujitsu.com.