Optimizing Business Location Decisions Using Machine Learning: An In-Depth Analysis

In the fast-paced world of business growth, picking the right location is essential to success. This study uses machine learning to help people in Cauayan City, Isabela, Philippines, make better decisions about where to put their businesses. The study's goal is to give investors useful information about the best places to start a business by combining demographic and risk data. The datasets are gathered from the data of the Local Government of Cauayan City, Isabela and Philippine National Police Station of the said City. It is a reputable sources during the data collection process, ensuring that they are relevant to the study area and up-to-date. The IBPAP Score Card was also utilized to identify promising cities for IT-BPM operations and to attract investment to this location. Supervised learning algorithms, like decision trees and random forests, are used in this method to sort barangays into groups based on their risk levels. K-means clustering and other unsupervised learning techniques are also used to find demographic patterns across the city. Crime, business, population, and graduate risk levels, as well as demographic distributions, are all included in the datasets that were used. The Decision Random Forest model's hyperparameter tuning leads to an 80% increase in accuracy, showing how well the model works at finding low-risk areas. Using feature importance analysis, we can see that crime and population risk levels are good indicators of business risk. This shows how important these factors are when making decisions. The model does a good job of putting areas with low risk into groups, In contrast, the Gradient Boosting model demonstrated exceptional performance in the prediction of medium and high-risk categories, achieving a perfect recall despite a lower overall accuracy. Additionally, identifying risk clusters with K-Means clustering helped allocate resources and target interventions. A further discovery made by the clustering analysis was the identification of three distinct risk clusters, each of which calls for a different intervention strategy. The significance of key predictors of business risk, including crime and population risk levels, was underscored in the context of strategic decision-making. The results show how machine learning could change business location strategies by giving people a data-driven way to make better investment decisions. This study makes it possible for businesses to plan their strategies and plans with more knowledge. This will help the economy grow and lower the risks of investing in Cauayan City.