EX

Enterprise AI Administration

High-Value Subscriber Segmentation Model

Model Details · Telecom AI Model Registry

Model Details

High-Value Subscriber Segmentation Model

Algorithm: Gaussian Mixture ModelAPPROVED DESIGN
Code
MDL-SEG-HV
Version
v0.1
Owner
Customer Analytics
Governance owner
AI Governance Board

Governance configuration shown for review and validation. Read-only.

Business purpose

Identify high-value subscriber groups with probabilistic membership

Production-recommended design specification. This record does not mean the model has been trained, fitted, deployed or validated.

Technical specification

Model family
Probabilistic Clustering
Algorithm
Gaussian Mixture Model
Implementation library
GaussianMixture
Model type
Unsupervised Probabilistic Clustering
Explainability method
Cluster profile; membership probability
Cluster selection method
BIC / AIC / validation during model development
Distance / membership method
Probabilistic cluster membership

Implementation governance

Governance status
APPROVED DESIGN
Implementation status
MODEL DEVELOPMENT REQUIRED
Validation status
VALIDATION REQUIRED
Lifecycle stage
Design
Current performance
Not yet measured
Refresh cadence
Monthly
Last reviewed
—
Next review due
—

Business

Mission area
Revenue & ARPU
Decision journey
Revenue & Profitability
Target variable
—
Output
Segment membership probabilities

Input features

Governed variables, in feature-rank order

  1. 1.average_monthly_revenueAverage billed service revenue over the last 3 months.
  2. 2.Plan typePrepaid, hybrid or postpaid
  3. 3.Tenure monthsMonths since activation
  4. 4.Payment timelinessShare of postpaid bills paid on time
  5. 5.churn_propensityChurn Propensity Model score (0–1).

Related Intelligence

Linked governed records

Decision journey