AI-Driven E-Commerce Pricing: Sentiment Analysis, Deep Learning & Reinforcement Learning Models

The integration of Artificial Intelligence (AI) in e-commerce has revolutionized pricing strategies by enabling dynamic adjustments based on customer sentiment and market trends. Traditional pricing models, reliant on fixed rules, often fail to account for consumer sentiment, leading to inefficiencies. This study proposes an AI-powered dynamic pricing framework that combines sentiment analysis with deep learning (LSTM, ANN) and reinforcement learning (PPO) to optimize pricing strategies dynamically. Customer sentiment, extracted from product ratings, is integrated into pricing decisions. An Artificial Neural Network (ANN) ensures adaptive and stable price modifications, while Long Short-Term Memory (LSTM) predicts future price fluctuations using historical data. Proximal Policy Optimization (PPO) fine-tunes prices in real-time, learning from market responses to maximize revenue. Experimental results demonstrate that AI-based pricing significantly influences price distributions, with ANN providing balanced adjustments, LSTM excelling in demand forecasting, and PPO offering dynamic adaptability for real-time optimization. Additionally, the study introduces an AI-powered marketing strategy recommendation system to align promotional efforts with sentiment trends, enhancing consumer engagement and profitability. The findings underscore the importance of sentiment-aware AI pricing models in modern e-commerce, offering a scalable and robust framework for revenue generation and market competitiveness.