Predicting Charge Off Using Text Analysis
Presentation Type
Poster
Student
Yes
Track
Finance/Insurance Application
Abstract
This research attempts to predict charge off for loans based on an open text box filled out by the loan applicant to encourage investors to invest in their loans. To predict charge off, latent sematic analysis is used with classification tools, techniques such as random forest, singular value decomposition, and tokenization, and TF-IDF. Preliminary exploratory data analysis shows that text length does not seem to have a strong association with charge off.
Start Date
2-5-2019 12:00 PM
End Date
2-5-2019 1:00 PM
Predicting Charge Off Using Text Analysis
Volstorff A
This research attempts to predict charge off for loans based on an open text box filled out by the loan applicant to encourage investors to invest in their loans. To predict charge off, latent sematic analysis is used with classification tools, techniques such as random forest, singular value decomposition, and tokenization, and TF-IDF. Preliminary exploratory data analysis shows that text length does not seem to have a strong association with charge off.