Data Combination for Problem-solving: A Case of an Open Data Exchange Platform

TL;DR

This study investigates how combining diverse datasets from different domains can effectively contribute to problem-solving, revealing that even small or non-overlapping datasets can be valuable in practical applications.

Contribution

It provides empirical evidence on the characteristics of data that facilitate problem-solving through combination, especially in large-scale and interdisciplinary contexts.

Findings

Small datasets are frequently used in solutions.

Non-overlapping datasets can still be effective.

Data combination mechanisms enable problem-solving with diverse data.

Abstract

In recent years, rather than enclosing data within a single organization, exchanging and combining data from different domains has become an emerging practice. Many studies have discussed the economic and utility value of data and data exchange, but the characteristics of data that contribute to problem solving through data combination have not been fully understood. In big data and interdisciplinary data combinations, large-scale data with many variables are expected to be used, and value is expected to be created by combining data as much as possible. In this study, we conduct three experiments to investigate the characteristics of data, focusing on the relationships between data combinations and variables in each dataset, using empirical data shared by the local government. The results indicate that even datasets that have a few variables are frequently used to propose solutions for problem solving. Moreover, we found that even if the datasets in the solution do not have common variables, there are some well-established solutions to the problems. The findings of this study shed light on mechanisms behind data combination for problem-solving involving multiple datasets and variables.