Publicación: MODELO PREDICTIVO DE FUGA DE CLIENTES UTILIZANDO MINERÍA DE DATOS PARA UNA EMPRESA DE TELECOMUNICACIONES EN CHILE
Fecha
2014
Título de la revista
ISSN de la revista
Título del volumen
Editor
UNIVERSIDAD, CIENCIA Y TECNOLOGÍA
Resumen
THIS WORKS SHOWS AN APPLICATION OF A CHURN MODEL FOR A COMPANY OF THE TELECOMMUNICATIONS INDUSTRY IN CHILE WITH TWO MAJOR MARKETS. CLUSTER ANALYSIS AND MULTIVARIATE LOGISTIC REGRESSION WERE USED AS DATA MINING METHODOLOGIES TO GENERATE THE CHURN MODEL. DATA BASE CONSIDERED THE PERIOD JANUARY 2009 AND JUNE 2011, AND VARIABLES SUCH AS PRODUCTS, DEMOGRAPHIC DATA, PAYMENT SYSTEMS, NUMBER AND TYPES OF COMPLAINTS FILLED WERE USED. TWO MODELS USING REGRESSION ANALYSIS WERE DEVELOPED WERE GENERATED FOR BOTH MARKETS. FOR THE CONCEPCIÓN MARKET THE MODEL ESTIMATES SUCCESSFULLY 71% OF THE TOTAL CASES, AND FOR TEMUCO MARKET WAS 72%. THE ABOVE INFORMATIONS ALLOWS THE COMPANY FOR THE PROPOSAL OF PROMOTIONAL EFFORTS AND TRAINING OF THE SALES FORCE PRIMARILY
Descripción
Palabras clave
REGRESIÓN LOGÍSTICA MULTIVARIABLE, MINERÍA DE DATOS, FUGA DE CLIENTES, ANÁLISIS DE GRUPOS
