Predictive Model Building for Driver-Based Budgeting Using Machine Learning

Abstract

Budgeting in the traditional sense is simply too slow and rigid to keep pace with the swiftly changing business environment. At the moment, there is far too much volatility, complexity, and uncertainty. A driver-based planning and budgeting model is more data-driven than a traditional budget model. This budgeting strategy shortens the time it takes to create a budget. Most driver-based planning and budgeting models center on predictions. One of the most difficult aspects of using driver-based planning, however, is identifying appropriate business drivers and predicting the impact of these drivers. Machine learning can assist driver-based budgeting processes in identifying the key drivers and predicting the impacts of these drivers. This study discusses the building of predictive modeling using machine learning. It illustrates stages from quantifying the budgeting issues to determining the best predictive mode for driver-based budgeting.

Sentiment Classification of Community towards COVID-19 Issues on Twitter (Case Study: Indonesia, March-May 2020)

“…Machine learning is a branch of artificial intelligence that allows computers to "learn" from data and develop the ability to make predictions or decisions without being explicitly programmed ( [58], [59]). The main advantage of machine learning is its ability to handle complexity and uncertainty in text data, process large volumes of data efficiently, and automatically update its model to improve accuracy ( [60], [61]).…”