Big data analysis of public library operations and services by using the Chernoff face method

Abstract

Purpose

The purpose of this paper is to conduct a big data analysis of public library operations and services of two cities in two countries by using the Chernoff face method.

Design/methodology/approach

The study is designed to evaluate library services by analyzing the Chernoff face. Big data on public libraries in London and Seoul were collected, respectively, from Chartered Institute of Public Finance and Accountancy and the Korean government’s website for drawing a Chernoff face. The association of variables and human facial features was decided by survey. Although limited in its capacity to handle a large number of variables (eight were analyzed in this study) the Chernoff face method does readily allow for the comparison of a large number of instances of analysis. A total of 58 Chernoff faces were drawn from the formatted data by using the R programming language.

Findings

The study reveals that most of the local governments in London perform better than those of Seoul. This consequence is due to the fact that local governments in London operate more libraries, invest more budgets, allocate more staff and hold more collections than local governments in Seoul. This administration resulted in more use of libraries in London than Seoul. The study validates the benefit of using the Chernoff face method for big data analysis of library services.

Practical implications

The Chernoff face method for big data analysis offers a new evaluation technique for library services and provides insights that may not be as readily apparent and discernible using more traditional analytical methods.

Originality/value

This study is the first to use the Chernoff face method for big data analysis of library services in library and information research.

Citation

Kim, Y.-s. and Cooke, L. (2017), "Big data analysis of public library operations and services by using the Chernoff face method", Journal of Documentation, Vol. 73 No. 3, pp. 466-480. https://doi.org/10.1108/JD-08-2016-0098