What is it about?
In this paper, we use some of the most popular clustering algorithms on a set of book titles from Early Modern Europe to compare how well they can clean the data. Clustering is a tool well suited for this, as the data we have, provided by OCLC, contains many duplicates with only minor changes, as well as different, but still valid, spellings of words that wouldn't be normalized by a spell checker. By cleaning and normalizing our data, we can improve the results we receive when querying the data in the future.
Featured Image
Photo by Luke Chesser on Unsplash
Why is it important?
Cleaning dirty data is a constant struggle in the data sciences. To properly be able to get information about our data set, we need to ensure it is as clean and normalized as possible, so data points don't slip through the cracks when we query the data set. This goes not only for us, but anybody working with large, imperfect data. By analyzing how each algorithm compares, we can help other researchers in this position choose the best algorithm for their data.
Read the Original
This page is a summary of: Analysis of Clustering Algorithms to Clean and Normalize Early Modern European Book Titles, January 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3451471.3451489.
You can read the full text:
Contributors
The following have contributed to this page
