Social media mining for business inequality: modified spotted hyena optimization based random forest

Abstract

Businesses may advertise, communicate with customers and research industry trends on social media today. Due to the massive amount of data generated on social media platforms, organizations struggle to acquire valuable data and fix marketing inequalities. To combine Modified Spotted Hyena Optimization with Random Forest (MSHO-RF) for classification in social media mining for a business inequality study. Data-driven solutions to inequality in operations and customer interactions should increase prediction accuracy, lower feature dimensionality and extract actionable insights. We begin by collecting social media mining for business inequality data. Afterward, Adaptive mean filter (AMF) techniques were used to initialize the data preprocessing. With the use of Linear Discriminant Analysis (LDA), we examined the most suitable arrangement of feature extraction. This research offers a unique method MSHO-RF, to deal with the problem of business inequality using social media mining. The MSHO-RF algorithm had higher results than compared to a variety of performance parameters, featuring a 94% of F1-score, a precision rate of 97%, an accuracy rate of 98% and recall rates of 95%. Our findings suggest that firms might benefit from a framework for leveraging social media data to promote more equitable and sustainable practices if they combine MSHO-RF with social media mining for business inequality. This approach benefits companies and their stakeholders by fostering an open, data-driven ecosystem.