System for predicting customer churn in telecommunication industry and subscription-based services, has prediction module for employing random forest algorithm to classify customer demographic, behavioral and transactional data into key churn predictors
	AGGARWAL R; KUMAR A; SINGH S; SOAM A; KUMAR L
	2025-03-30
专利权人	AGGARWAL R (AGGA-Individual) ; KUMAR A (KUMA-Individual) ; SINGH S (SING-Individual) ; SOAM A (SOAM-Individual) ; KUMAR L (KUMA-Individual) ; MEERUT ENG & TECHNOLOGY INST (MEER-Non-standard)
申请日期	2025-03-30
专利号	IN202511031431-A
成果简介	NOVELTY - The system has a data collection module that is configured to gather customer demographic, behavioral, and transactional data. A pre-processing module cleans and scales the data. A feature extraction module identifies key churn predictors. A prediction module employs a random forest algorithm to classify customers as likely to churn or retain with a confidence score. A user interface displays predictions, feature importance and retention recommendations. The pre-processing module performs one-hot encoding on categorical variables and standardization on numerical features and monthly charges, total charges and senior citizen status ranks features. USE - System for predicting customer churn in telecommunication industry and subscription-based services. ADVANTAGE - The system develops an accurate and reliable machine learning-based system for predicting customer churn, thus enabling businesses to identify at-risk customers with high precision using historical and real-time data. The system provides actionable insights into key factors driving churn, such as cost dissatisfaction or service quality, through feature importance analysis and visualizations, thus facilitating targeted retention strategies. The user interface provides input screens for data entry, prediction screens with results and confidence levels, and insight screens with visualizations such as bar charts and heatmaps, thus collectively enabling the business to identify churn drivers, simulate retention strategies, and reduce customer attrition effectively. DETAILED DESCRIPTION - An INDEPENDENT CLAIM is included for a method for predicting customer churn. DESCRIPTION OF DRAWING(S) - The drawing shows a schematic view of a system for predicting customer churn in telecommunication industry and subscription-based services.
IPC 分类号	G06Q-010/00 ; G06Q-030/0201 ; G06Q-030/0202
国家	印度
专业领域	信息技术
语种	英语
成果类型	专利
文献类型	科技成果
条目标识符	http://119.78.100.226:8889/handle/3KE4DYBR/13323
专题	中国科学院新疆生态与地理研究所
作者单位	1.AGGARWAL R (AGGA-Individual) 2.KUMAR A (KUMA-Individual) 3.SINGH S (SING-Individual) 4.SOAM A (SOAM-Individual) 5.KUMAR L (KUMA-Individual) 6.MEERUT ENG & TECHNOLOGY INST (MEER-Non-standard)
推荐引用方式 GB/T 7714	AGGARWAL R,KUMAR A,SINGH S,et al. System for predicting customer churn in telecommunication industry and subscription-based services, has prediction module for employing random forest algorithm to classify customer demographic, behavioral and transactional data into key churn predictors. IN202511031431-A[P]. 2025.

条目包含的文件
条目无相关文件。

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。