The use of clustering algorithms to prevent future office-related MSDs

• Main Author: Castro, Roselle Franchesca B.
• Format: text
• Language: English
• Published: Animo Repository 2021
• Subjects: Computer algorithms, Machine learning; Computer Sciences; Theory and Algorithms
• Online Access: https://animorepository.dlsu.edu.ph/etdm_infotech/3; https://animorepository.dlsu.edu.ph/cgi/viewcontent.cgi?article=1000&context=etdm_infotech
• Institution: De La Salle University

Computer-related injuries such as repetitive strain injuries and MSD is a growing issue yet is of less concern. This study focused on exploring the possibilities of the use of unsupervised machine learning techniques called clustering algorithms to prevent future computer-related injuries. Specifically using algorithms, KMeans, Hierarchical and DBSCAN. In exploring the concept of ergonomics, it then gave the understanding of the interaction between the human and the computers. It paved the way for the identification of the causes of computer-related injuries and prevention through an interview with an ergonomic expert. It was then used as the factors for the study to be explored. Feeding the factors to the three unsupervised clustering algorithms, resulted in the discovery of new groups that share the same characteristics, which became the basis to tell if a group may be the most at risk. Characterization provided valuable insight and possible action plans that will aid in prioritizing and mitigation plans of decision-makers. Execution will now be more tailored fit to the needs of the clusters. This is a step further an improvement from the current safety initiatives of an organization that aims to continuously improve the health and safety of the employees. This then explains how the use of clustering algorithms is a step further to help prevent future computer-related injuries. Keywords: MSD, Clustering Algorithm, Kmeans, DBSCAN, Hierarchical, unsupervised machine learning, analytics, preventive analytics