Global Liquor Insight Ensemble (GLIE) Algorithm: Big Data Analytics for Predicting Global Market Acceptance of Liquor Culture

Abstract

The use and acceptance of liquor culture varies greatly across worldwide marketplaces, owing to social, cultural, and financial factors. Comprehending these dynamics necessitates using big data analytic tools to identify consumer trends and desires. The purpose of this study is to use sophisticated machine learning and deep learning models to investigate and forecast global liquor usage trends, desires, and behaviours. The study aims to discover key characteristics and patterns that influence the market acceptability of liquor culture. Despite the diversity of liquor desires and usage practices, previous works lack thorough analytics that integrates big data to present meaningful insights into worldwide market dynamics. These drawbacks include ineffective handling of complicated customer buying patterns and poor forecasting performance. To overcome these limitations, this study introduces the GlobalLiquorInsightEnsemble (GLIE) Algorithm, which is intended to improve prediction accuracy and present deeper insights into liquor usage patterns. The study makes use of a dataset that includes demographic data, drinking behaviours, and liquor usage desires. The GLIE Algorithm includes ensemble machine learning models comprising REPTree, JRip, and Naive Bayes, as well as deep learning with DL4JMLPClassifier, for classification and prediction problems. Model evaluation measures include accuracy, precision, recall, f1-score, and Matthew's correlation coefficient (MCC). The study uses thorough analysis to identify major changes in liquor usage, preferences for certain types and Flavors of liquor, and patterns of behaviour connected with intake frequency and purchase channels. The ensemble models do well in forecasting customer behaviour across multiple global locations. Experimental results indicate that the suggested GLIE Algorithm attains an Accuracy of 91.1%, Precision of 90.5%, Recall of 89.3%, F1-score of 89.9%, and MCC of 81%, surpassing previous approaches and offering a more accurate and comprehensive understanding of global liquor consumption patterns.