Blockchain in churn prediction based telecommunication system on climatic weather application

• Published in: Sustainable computing informatics and systems Vol. 35; p. 100705
• Main Authors: Quasim, Mohammad Tabrez, Sulaiman, Adel, Shaikh, Asadullah, Younus, Mohammed
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.09.2022
• Subjects: Blockchain technology; Churn prediction; Classifier; Climatic Weather; Support vector machine; Telecommunication system; Churn prediction; Classifier; Support vector machine; Telecommunication system; Blockchain technology; Climatic Weather
• ISSN: 2210-5379
• DOI: 10.1016/j.suscom.2022.100705

For better customer service and customer retention, businesses take proactive measures, including troubleshooting and solving potential challenges promptly. Blockchain technology integrates different recognition techniques of distributed pattern for monitoring database of a dedicated network and has proven a very promising technology. An automated pattern recognition decentralizes and distributes customized specific data. Machine learning, originated from artificial intelligence is primarily related to recognition of behavior patterns. Methods like knowledge discovery in database (KDD) and data mining focus on unsupervised approaches and are widely used in business and climatic weather applications. Blockchain addresses data-security concerns and builds trust by creating distributed ledger. Theft, willful fraud, software and hardware are considered by blockchain in data protection. Blockchain has greater significant feature as it makes the secure data access without enabling the central management entity. Two design features of blockchain technology help in this task. Recommended customer data pattern recognition technique using blockchain may eliminate all these problems. Two kinds of cryptographic algorithms employed in blockchain are asymmetric-key algorithms and hash functions. The current study analyzes the asymmetric cryptography approach along with key pair which supports in system security. Recurrent neural network (RNN) and support vector machine (SVM) classifier techniques consider both old customers and new customers as stable. The predictive model aids in identifying customers at churn risk in the telecommunication system. Existing proactive methods are unable to explain difficulties in customer interaction understanding and meeting their genuine needs. The proposed model organizes the customer situation and designs a customer proactive re-engagement over mobile-based telecommunication systems. Performance measures like churn prediction, classifier, confusion matrix, machine learning in the telecommunication system are used to evaluate and validate the results. •Designs a customer proactive re-engagement over weather climatic application.•To Utilize the data set by controlling and diminish the subscriber churn issue.•Ensure secure data access without enabling the central management entity based on block chain for weather climate.•Integrating the block chain and cognitive communication ensure the secure data access.•To optimize and predict the cost and power by validating the proposed model based on churn factors and cluster profiling in climatic weather.