Real-time Consumer Behaviour Segmentation Using Transformer-based AI Models
Alurwad Tripat Venkatreddy
Government Degree College, Nirmal, Telangana, India.
K Krunal Yadav
Government Degree College (Arts and Commerce), Adilabad, Telangana, India.
Narote Preetham
Telangana Tribal Welfare Residential Degree College (Boys), Boath@Adilabad, Telangana, India.
Gajjala Lilly Rani
Avanthi’s Scientific Technological & Research Academy, Hyderabad, Telangana, India.
Ankatwar Gajanan
*
Government Degree College (Arts and Commerce), Adilabad, Telangana, India.
*Author to whom correspondence should be addressed.
Abstract
The rapid growth of digital commerce and online consumer interactions has generated vast amounts of behavioural data, creating new opportunities for organizations to understand customer preferences and improve marketing strategies. Consumer behaviour segmentation is a fundamental technique used to classify customers into distinct groups based on their characteristics and purchasing patterns. However, traditional segmentation methods such as K-Means clustering, RFM analysis, and conventional machine learning models often struggle to capture the dynamic and sequential nature of consumer behaviour in real-time environments. To address these limitations, this study proposes a Transformer-based Artificial Intelligence (AI) framework for real-time consumer behaviour segmentation.
The proposed framework utilizes self-attention mechanisms to analyse sequential consumer activities, including clicks, searches, cart additions, purchases, and reviews. By learning contextual relationships among behavioural events, the Transformer model effectively captures long-term dependencies and generates dynamic customer segments. The study employs a quantitative experimental research design using consumer behaviour datasets collected from e-commerce platforms. The performance of the Transformer model is compared with traditional machine learning and deep learning approaches, including K-Means, Random Forest, Support Vector Machine (SVM), and Long Short-Term Memory (LSTM) networks.
Experimental results demonstrate that the Transformer-based model achieves superior performance in terms of accuracy, precision, recall, and F1-score. The confusion matrix analysis further confirms the effectiveness of the proposed model in accurately classifying consumers into different behavioural segments. Additionally, real-time segmentation contributes to improved business outcomes, including higher customer retention, enhanced personalization, increased conversion rates, and better marketing effectiveness. The findings highlight the potential of Transformer-based AI models as a scalable and intelligent solution for next-generation customer analytics and real-time decision-making systems.
Keywords: Consumer behaviour, real-time segmentation, transformers, deep learning, artificial intelligence, customer analytics, e-commerce