數據分析 Journey Phase 3

After the modern data platform has been well-implemented, we can enhance the analytics with ML (機器學習 (ML)) to get more insights.

例如

1. Customer segmentation by ML
2. 設立評分系統
3. 推薦引擎

根據客戶評分和細分

客戶評分基於客戶數據庫的細分

Customer scoring draws on customer data. The score assigned to each of your customers is calculated from the data you have on them. You can use all types of data to build a scoring system:

• Socio-demographic data: age, sex, marital status, profession
• Psychological data: interests, opinions
• Behavioral data: purchase history, data of last purchase, purchase frequency, Apps behavior, Campaign conversion rate, etc.

細分示例

Segmentation by 機器學習 (ML)

找到四類型的客戶

1. 年收入低且支出得分低（智能購物者是銷售/優惠券/促銷的高度針對性對象）
2. 年收入低且支出得分高（滿意的購物者由於他們的消費習慣而對其最不感興趣）
3. Average income with Average spending score (Require more data to figure out their buying decision)
4. High income with Low spending score (Unsatisfied shopper with the mall’s service. These are our target study group as we need to attract theses shoppers to increase sales demand)

推薦引擎

Remarks: Location data and Push Notification action will be handled by a 3rd party vendor.

Case Reference: Delivery Optimization  (Chain store scheduler)

Example usages of 機器學習 (ML):

• 客戶評分
• 客戶細分
• 客戶行為分析
• 數據 Profiling
• 異常檢測

#ML #MachineLearning #DeepLearning #AI #ArtificialIntelligence #ModernDataPlatform #DataConsultation #DataPlanning #DataModeling #DataAnalyticsJourney #DataJourney #HongKongDataAnalytics #HongKongDataAnalysis #PredictiveAnalysis #ForecastAnalysis #TrendAnalysis #Azure #PowerBI #CloudMigration