The credit score is a numeric value measuring people’s creditworthiness and used by banks to assess customer credit applications. It is common practice to use variables like monthly income, debt ratio, open credit loans, demographics etc with supervised machine learning approaches. Based on your experience what are the best practices in terms of feature engineering and algorithms? Do you use unsupervised and deep learning approaches for credit scoring and how?
Dairi Abdelkader
Dear,
I suggest to take a look on Restricted Boltzmann Machine, Deep Belief Network both are probabilistic model trying to learn data distribution of the input (training dataset) regardless the data type, the training is entirely unsupervised
0 votes
0 thanks
- Badges
- Science topic
More Deger Ayata's questions See All
Similar questions and discussions