customer_sales_churn_small

This dataset contains sales and behavioral data for 10,000 customers, primarily designed for churn analysis and customer segmentation. It includes metrics tracked over several weeks (labeled W1 through W5) to capture trends in purchasing behavior.

Key Components of the Dataset:
Customer Identification: CUSTOMER_ID uniquely identifies each record.

Sales Metrics: Includes total sales (Total_Sale), standard deviation of sales (STD_Sales), and average purchase value (APV). Note that many sales values appear to be log-transformed or normalized.

Temporal Tracking (Weekly Data): Detailed breakdown for weeks 1–5, including:

Visits: Number of visits per week (W1_Visits, etc.).

Sales Performance: Minimum, maximum, and standard deviation of sales per week (e.g., W3_Max_Sale, W4_Min_Sale).

Activity Flags: Binary indicators for whether a customer was active in specific weeks (week_1, week_2, etc.).

Recency: Days_since_last_visit tracks how long it has been since the customer's last interaction.

Customer Segmentation:

Value-based: Categorizes customers as High_value, Low_value, or Regular.

Engagement-based: Categorizes customers by visit frequency (e.g., Frequent_Visitors, Rare_Visitors).

Target Variable:

CHURN: A binary indicator (0 or 1) typically used to predict if a customer will stop using the service.

原始信息汇总

数据集概述

基本信息

• 数据集名称: customer_sales_churn_small
• 数据集地址: https://bigml.com/user/jarcosr/gallery/dataset/6978e22e4a287edd13a47127
• 创建时间: Tue, 27 Jan 2026 16:05:02 +0000
• 发布时间: Tue, 27 Jan 2026 16:14:35 +0000
• 数据集大小: 2.8 MB
• 字段数量: 36
• 实例数量: 10,000

数据集描述

此数据集包含10,000名客户的销售和行为数据，主要用于流失分析和客户细分。它包含在数周（标记为W1至W5）内跟踪的指标，以捕捉购买行为趋势。

数据集关键组成部分

• 客户识别: CUSTOMER_ID唯一标识每条记录。
• 销售指标: 包括总销售额（Total_Sale）、销售标准差（STD_Sales）和平均购买价值（APV）。许多销售值似乎是经过对数转换或标准化的。
• 时间跟踪（周数据）: 第1至5周的详细细分，包括：
  • 访问次数: 每周访问次数（W1_Visits等）。
  • 销售表现: 每周销售的最小值、最大值和标准差（例如W3_Max_Sale、W4_Min_Sale）。
  • 活动标志: 客户在特定周（week_1、week_2等）是否活动的二进制指标。
• 最近度: Days_since_last_visit跟踪自客户上次互动以来的天数。
• 客户细分:
  • 基于价值: 将客户分类为高价值（High_value）、低价值（Low_value）或常规（Regular）。
  • 基于参与度: 按访问频率对客户进行分类（例如Frequent_Visitors、Rare_Visitors）。
• 目标变量: CHURN是一个二进制指标（0或1），通常用于预测客户是否会停止使用服务。

字段示例（部分）

字段名	类型	计数	缺失值	错误
CUSTOMER_ID		10,000	0	0
Total_Sale		10,000	0	0
STD_Sales		10,000	0	0
Hist_Visits		10,000	0	0
W1_Min_Sale		10,000	0	0
W1_STD_Sales		10,000	0	0
W1_Visits		10,000	0	0
W2_Min_Sale		10,000	0	0
W2_STD_Sales		10,000	0	0
W2_Visits		10,000	0	0

标签

• 所有者: jarcosr
• 状态: Public
• 获取方式: FREE