辽宁省淘宝平台食品类客户分级评价数据

采集辽宁省淘宝平台客户消费行为数据，通过客户的最近一次消费时间（R）、最近一段时间消费频次(F)、最近一段时间消费金额(M)，采用RFM模型对客户进行价值评级，实现精准化运营，通过对辽宁省淘宝平台客户价值管理，满足不同价值客户的个性化需求。并为同行业企业不同价值类型的客户个性化服务提供数据支持。1、数据处理：对采集到的数据进行降噪、清洗、脱敏、聚集、分析。
2、数据加工：运用RFM模型结合用户的最近一次活动（R）、用户活动频率（F）和消费金额（M）的得分排名对客户进行一个综合排名，最终得出一个RFM总评分。a.提取出最近一次消费时间（R）、最近一段时间消费频次(F)、最近一段时间消费金额(M)，将用户按照最近一次活动（R）进行分类，最近一次活动时间间隔最短的用户排在最上面。按照从1-5评分，前20%的客户获得5分，接下来的20%用户获得4分,再下来20%的客户为3分，再下来20% 的客户为2分，最后20% 的客户为1分。b.根据客户活动频率（F）从高到底依次对用户进行分类，前20%的客户在用户活动频率的分数为5，以此类推。c.消费金额（M），前20%的客户在消费金额的分数为5，以此类推。消费金额最少的20%客户则分数为1。RFM得分=（R）得分*0.3+（F）得分*0.3+（M）得分*0.4。评分大于等于4分的为A级客户，大于等于3小于4的为B级客户，大于等于2小于3的为C 级客户，低于2的为D 级客户。
3、通过对客户的分级管理，为不同价值类型的客户个性化服务提供数据支持。

This dataset collects consumer behavior data of Taobao platform customers in Liaoning Province. Utilizing the RFM model— which incorporates three key indicators: Recency (R, the time elapsed since the customer’s most recent consumption), Frequency (F, the total number of consumption transactions within a specified recent period), and Monetary value (M, the total consumption amount within the same recent period)— this work conducts customer value segmentation to enable precise operational optimization. Through targeted customer value management for Taobao users in Liaoning Province, the dataset meets the personalized needs of customers across different value tiers, and offers data support for peer enterprises to develop personalized services for their customer groups with distinct value types.
1. Data Preprocessing: Denoise, clean, anonymize, aggregate and analyze the collected raw consumer behavior data.
2. RFM-based Scoring and Ranking: Apply the RFM model to generate a comprehensive ranking of customers based on the score rankings of their Recency, Frequency and Monetary value metrics, and finally calculate an overall RFM total score.
a. Extract the three core indicators: R (time since last consumption), F (consumption frequency in recent period) and M (consumption amount in recent period). Classify customers by their R metric: customers with the shortest time interval since their last consumption are ranked highest. Assign scores from 1 to 5 using quintile segmentation: the top 20% of customers receive 5 points, the next 20% get 4 points, the subsequent 20% get 3 points, the following 20% get 2 points, and the bottom 20% receive 1 point.
b. Classify customers in descending order of their F metric: the top 20% are awarded 5 points, with the remaining customers assigned scores following the same quintile rule.
c. For the M metric, the top 20% of customers by total consumption amount get 5 points, while the bottom 20% (with the lowest total consumption) receive 1 point, following the same quintile scoring logic.
The overall RFM score is computed as: RFM Score = (R Score) × 0.3 + (F Score) × 0.3 + (M Score) × 0.4. Customers with a score ≥4 are categorized as Level A customers, those with 3 ≤ Score <4 as Level B customers, those with 2 ≤ Score <3 as Level C customers, and those with Score <2 as Level D customers.
3. Through hierarchical customer value management, this dataset provides data support for enterprises to deliver personalized services tailored to customers of different value types.