Hybrid Machine Learning and Big Data Analytics Framework for Predictive Business Intelligence and Data-Driven Decision-Making

Abstract

The business data production has surged significantly due to the use of digital platforms, enterprise applications and transactions being conducted over the Internet resulted in the huge demand for the advanced analytical tools in order to apply predictive business intelligence and the data-driven decision-making. The traditional business intelligence software is not capable of processing large amounts of heterogeneous data, recognizing complex patterns in data, predicting business trends and providing timely insights. In order to solve these problems, this paper suggests a new Hybrid Machine Learning and Big Data Analytics Framework as the Predictive Business Intelligence Tool. The suggested framework combines distributed big data processing, data preprocessing and modeling, feature engineering, machine learning algorithms, and analytical visualization which allows transforming complex business data into the useful insights. The proposed framework delivers a number of functions such as demand forecasting, consumer behavior evaluation, risk forecasting, anomaly detection, and business effectiveness improvement. Its hybrid architecture makes it possible to bring together the scalable data processing technologies and predictive analytics algorithms making the analytics process even more efficient. The proposed framework can be assessed in terms of accuracy, precision, recall, F1 score, root mean squared error among others depending on the type of the task to perform. The framework will allow providing efficient and adaptive analytics processes within a variety of business areas including retail, finance, production, and supply chain management.

Introduction

The rapid development of digital technologies, artificial intelligence (AI), cloud computing, and big data analytics has significantly increased the volume and complexity of business data. Traditional Business Intelligence (BI) systems mainly focus on historical reporting and descriptive analysis, limiting their ability to predict future trends and support complex decision-making. To overcome these limitations, this study proposes a hybrid framework that integrates Machine Learning (ML) techniques with big data analytics to improve predictive business intelligence and data-driven decision-making.

The proposed framework combines data processing, feature extraction, predictive modelling, and visualization to convert large volumes of structured and unstructured business data into meaningful insights. It uses various machine learning algorithms, including Random Forest, Gradient Boosting, Support Vector Machines (SVM), regression techniques, and clustering methods, depending on the business problem. The framework supports applications such as sales forecasting, customer behaviour analysis, customer segmentation, risk assessment, anomaly detection, inventory management, and operational optimization.

The methodology consists of four major stages:

1. Business data acquisition and preprocessing: Data is collected from sources such as customer relationship management systems, transaction records, online sales platforms, market research, and supply chains. Data cleaning, duplicate removal, missing-value handling, normalization, and data integration are performed to improve data quality.

2. Feature extraction and preparation: Relevant variables, such as sales history, customer purchasing patterns, product demand, pricing, inventory, and seasonal trends, are selected and transformed into suitable features. The dataset is divided into training, validation, and testing sets for reliable model development and evaluation.

3. Hybrid machine learning and predictive analytics: Multiple ML algorithms are applied to identify patterns, forecast future outcomes, classify business risks, and discover hidden relationships. Model selection and hyperparameter tuning are used to improve predictive performance.

4. Insight generation and decision support: Predictions are presented through reports, dashboards, and visualizations to support business planning, resource allocation, inventory optimization, targeted marketing, and risk management. Models can also be retrained periodically using new data to maintain their performance.

The literature review highlights the growing importance of integrating AI, ML, and big data analytics into business intelligence systems. However, practical implementation remains challenging because of data inconsistencies, missing information, computational costs, class imbalance, and changing market conditions. Therefore, the proposed framework emphasizes scalable data processing, continuous performance evaluation, and adaptability to different business environments.

The reported experimental results indicate that the proposed hybrid framework outperforms the individual machine learning models included in the comparison.

• Classification performance: The hybrid framework achieved 93.26% accuracy, 92.91% precision, 92.74% recall, and a 92.81% F1-score, outperforming Logistic Regression, Random Forest, Support Vector Machine, and Gradient Boosting.

• Forecasting performance: The framework achieved a mean absolute error (MAE) of 6.48, mean squared error (MSE) of 82.16, and root mean square error (RMSE) of 9.06, showing lower prediction errors than the other forecasting methods listed.

• Evaluation criteria: The framework considers predictive accuracy, precision, recall, F1-score, forecasting errors, computational efficiency, memory usage, and scalability to assess its effectiveness.

Conclusion

The results of the illustration give the classification precision of 93.26 ± 0.42%, the F1 score of 92.81 ± 0.43%, and the forecasting RMSE of 9.06 ± 0.48. The illustrative RMSE shows a decrease in the forecasting error by 16.63% compared to that of the Gradient Boosting Regression. Besides, the structure shows the illustrative prediction latency of 3.76 ± 0.17 ms/sample and processing throughput of 10,650 ± 285 records/s. The evidences given prove that integrated machine learning and big data analytics can contribute to predictive analytics for business intelligence, resource allocation, and making informed decisions. The numerical values presented are illustrative and will be replaced with the data obtained in lab experiments. In the future, research will involve evaluating the structure using real conditions, such as massive real-time datasets from companies operating in the retail, finance, manufacturing, and logistics fields. The possibilities of advanced ensemble algorithms, deep learning, explainable AI, federated learning, and adaptive retraining can be researched to enhance prediction dependability, interpretability, privacy, and scalability.

References

[1] J. C. Ines et al., \"The Role of AI in Enhancing Decision-Making in Small and Medium Enterprises (SMEs),\" 2024 2nd International Conference on Computing and Data Analytics (ICCDA), Shinas, Oman, 2024, pp. 1-4, doi: 10.1109/ICCDA64887.2024.10867356. [2] U. Mangal, S. Mogha and S. Malik, \"Data-Driven Decision Making: Maximizing Insights Through Business Intelligence, Artificial Intelligence and Big Data Analytics,\" 2024 International Conference on Advances in Computing Research on Science Engineering and Technology (ACROSET), Indore, India, 2024, pp. 1-7, doi: 10.1109/ACROSET62108.2024.10743399. [3] A. Usmanov, A. Valiyev, M. Akhunova, B. Makhmudov and D. Uzganbayeva, \"AI-Powered Business Intelligence: Redefining Data-Driven Decision-Making,\" 2026 2nd International Conference on Sustainable Computing and Integrated Communication in Changing Landscape of AI (ICSCAI), Greater Noida, India, 2026, pp. 1491-1498, doi: 10.1109/ICSCAI68849.2026.11648682. [4] S. V. Khara, G. D. Tivari, S. Patel, J. Nathvani, R. Amrutiya and D. Chavda, \"Generative Artificial Intelligence for Intelligent Decision Support in Organizational Management: A Data-Driven Framework,\" 2026 3rd International Conference on Research Methodologies in Knowledge Management, Artificial Intelligence and Telecommunication Engineering (RMKMATE), Chennai, India, 2026, pp. 1-5, doi: 10.1109/RMKMATE69073.2026.11518863. [5] B. B. Khadka, S. Raj Bhandari, A. B. Rokaya, H. Jit Kaur and R. Markan, \"Data Driven Decision Making and Analytics in Contemporary Business Management,\" 2025 International Conference on Intelligent & Innovative Practices in Engineering & Management (IIPEM), Singapore, Singapore, 2025, pp. 1-5, doi: 10.1109/IIPEM65914.2025.11547928. [6] B. K. Sharma, J. J. Prakash, A. Dwivedi, V. Sharma, S. Singh and V. Singh, \"A Self-Healing 6G Network Model Using AI-Driven Predictive Analytics,\" 2026 International Conference on Computing, Communication, Control and Cyber-Physical Systems (I5CPS), Karaikal, Puducherry, India, 2026, pp. 1-7, doi: 10.1109/I5CPS67958.2026.11452521. [7] C. R. W. Robles, R. D. Palenzuela and R. C. Raga, \"Business Intelligence Integration and Data Analytics for Enhanced Hospitality Management in Neptune Travel and Tours,\" 2025 16th International Conference on E-Education, E-Business, E-Management and E-Learning (IC4e), Tokyo, Japan, 2025, pp. 185-190, doi: 10.1109/IC4e65071.2025.11075406. [8] P. Kumar and P. Dahiya, \"Sharpening Business Decision Making through Artificial Intelligence and Digital Marketing: An Empirical Study,\" 2025 2nd International Conference on Computational Intelligence and Computing Applications (ICCICA), Samalkha, India, 2025, pp. 1-6, doi: 10.1109/ICCICA67008.2025.11337492. [9] P. Usha and S. K. Puli, \"Machine Learning for Strategic Business Intelligence: A Short Review,\" 2024 IEEE International Conference on Computing, Applications and Systems (COMPAS), Cox\'s Bazar, Bangladesh, 2024, pp. 1-5, doi: 10.1109/COMPAS60761.2024.10796605. [10] V. Sharma, R. K. Mishra, S. Shrotriya, S. Kumar, V. Singh and R. Kant, \"Blockchain Integrated Intrusion Detection System for Advanced Cyber Physical Networks,\" 2026 2nd International Conference on Innovations in Mechanical, Robotics, Computing and Biomedical Engineering (ICIMRBE), ERNAKULAM, India, 2026, pp. 1-6, doi: 10.1109/ICIMRBE70745.2026.11709985. [11] S. Hundekari, A. Shrivastava, R. Praveen, R. H. C. Alfilh, A. Badhoutiya and N. Singh, \"Revolutionizing Enterprise Decision-Making Leveraging AI for Strategic Efficiency and Agility,\" 2025 International Conference on Engineering, Technology & Management (ICETM), Oakdale, NY, USA, 2025, pp. 1-6, doi: 10.1109/ICETM63734.2025.11051858. [12] R. Ravindran, R. Vettriselvan, P. R. Velmurugan, S. P N and M. Sakthivanitha, \"Leveraging Artificial Intelligence for Data-Driven Decision-Making in Business Organizations,\" 2026 5th International Conference on Sentiment Analysis and Deep Learning (ICSADL), Birendranagar, Nepal, 2026, pp. 1-9, doi: 10.1109/ICSADL67539.2026.11451857.

Copyright

Copyright © 2026 Abhishek Kumar Pathak , Dilip Kumar. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.