Enhancing Recruitment Efficiency in the Indonesian Pharmaceutical Industry Using Backpropagation Neural Networ

Enhancing Recruitment Efficiency in the Indonesian Pharmaceutical Industry Using Backpropagation Neural Networ

backpropagation; deep learning; employee selection model; multi-layer perception; recruitment

PT K-24 Indonesia is one of the foremost companies in Indonesia, with a primary focus on distributing pharmaceutical products and healthcare services. Over the last two years, PT K-24 has received more than 110,000 job applicants, offering various position vacancies. The recruitment process began with registrations, online tests, and interviews. The need for manpower has increased annually in recent years. More attention is required when selecting prospective employees who meet the selection criteria. However, PT K-24 found that the recruitment process was less efficient because the applicants did not meet the company’s criteria. To overcome this problem, it is necessary to create an employee selection model for candidate selection. This study developed a recommendation using the multilayer perceptron method, namely, backpropagation. The results of this study add to the confidence of prior research that the backpropagation algorithm is effective for prediction and reliable forsimilar prediction tasks. This study found that using 112,397 relevant data, 23 input parameters, 1 hidden layer containing 10 neurons, ReLu activation function, and Adam solver produced better prediction, with an MSE of 0.029. Our study contributes to the methodological approach by implementing real-world problems and measuring specific criteria relevant to the selection process.

Apfia Normalita1, Jong Jek Siang2*, Halim Budi Santoso

https://jurnal.iaii.or.id/index.php/RESTI/article/view/6303/1133

Department of Information Systems, Duta Wacana Christian University, Yogyakarta, Indonesia

October3, 2025

pdf

ENGLISH

TEXT