"Comparación de técnicas de clasificación en la recomendación de productos bancarios"
The present research is entitled "Comparison of Classification Techniques in the Recommendation of Banking Products", which addresses the generation of recommendations as a classification problema, comparing the classifiers provided by the Naïve Bayes (NB), Multi-layer Feed-forward Neural...
Bewaard in:
| Hoofdauteur: | |
|---|---|
| Formaat: | bachelorThesis |
| Taal: | spa |
| Gepubliceerd in: |
2017
|
| Onderwerpen: | |
| Online toegang: | http://repositorio.uteq.edu.ec/handle/43000/4067 |
| Tags: |
Voeg label toe
Geen labels, Wees de eerste die dit record labelt!
|
Voeg label toe | Samenvatting: | The present research is entitled "Comparison of Classification Techniques in the Recommendation of Banking Products", which addresses the generation of recommendations as a classification problema, comparing the classifiers provided by the Naïve Bayes (NB), Multi-layer Feed-forward Neural Networks (NNET), Support Vector Machines (SVM) and Multivariate Adaptive Regression Splines (MARS) algorithms, which were selected based on their applicability in the present classification problem; for this we used the AUC, training and test time, with the purpose of being able to determine an algorithm that provides classifiers with a high performance to generate recommendations of banking products. Keywords: Data mining, classification, recomendations. |
|---|