ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-04122019-183657


Tipo di tesi
Tesi di laurea magistrale
Autore
SHTJEFNI, MARIO
URN
etd-04122019-183657
Titolo
Using machine learning and linear programming approaches to optimize TV advertising campaign performance
Dipartimento
INFORMATICA
Corso di studi
INFORMATICA PER L'ECONOMIA E PER L'AZIENDA (BUSINESS INFORMATICS)
Relatori
relatore Prof. Pedreschi, Dino
Parole chiave
  • optimization
  • model interpretation
  • machine learning
  • linear programming
  • forecasting
  • spot booking
Data inizio appello
03/05/2019
Consultabilità
Completa
Riassunto
TV advertisers have a limited budget they can spend on a television ad campaign. When promoting a product or service, their goal is to optimally allocate spot bookings on TV channels so as to reach the biggest audience possible while at the same time not surpassing the financial resources associated with the campaign. In this paper, I have used machine learning models to analyze data of previous spot bookings and the corresponding consumer response to determine what features are the most important in predicting a successful campaign. Subsequently, based on model forecasts I have proposed an optimal basket of booking choices which takes all budgetary constraints into account and can be implemented by the advertiser for future campaigns.
File