个人贷款流失挽回预测模型数据集

个人贷款流失挽回预测模型是基于数据挖掘，特征工程，经过机器学习算法得出的概率型预测模型。模型目标为预测出在我行已结清贷款客户，在接下来的一段时间内是否会再次贷款的概率，并向业务人员进行推送。建模过程中分别选取了最近一次贷款、基本信息、历史贷款、存款与收单、征信、授信与用信等近600个特征，经过筛选后入模200余特征，使用梯度提升决策树算法对数据进行训练，使模型能够准确预测出可进行挽回的客户，该数据模型可以帮助客户经理提高营销效率。

The Personal Loan Attrition Recovery Prediction Model is a probabilistic predictive model developed via data mining, feature engineering, and machine learning algorithms. The model aims to predict the probability that a customer who has fully repaid their loan at our bank will apply for another loan within a specified future period, and push the prediction results to business personnel. During model development, nearly 600 features were extracted from multiple dimensions including the most recent loan, basic customer information, historical loan records, deposits and acquiring transactions, credit reports, credit granting and utilization, etc. After feature screening, over 200 features were selected for model training. Gradient Boosting Decision Tree (GBDT) algorithm was adopted to train the dataset, enabling the model to accurately identify customers eligible for recovery efforts. This model can help customer managers improve their marketing efficiency.