NPL DETECTION OF LARGE BUSINESS IN BANKING DOMAIN

by | Nov 16, 2021 | 0 comments

Application Track:

Ready Made

Code:

REACH-2021-READYMADE-YKT_1

Domain:

Proposed by:

YAPI KREDI TEKNOLOJİ (YKT)

Entity Logo:

Summary of the entity:

Yapı Kredi Teknoloji is a Turkish company that delivers outputs in machine learning, data mining, pattern recognition, artificial intelligence and natural language processing fields and develops mobile applications.

Summary of the challenge:

YKT aims to identify the large scale firms that are probable to be default through analysing their financial data.

Description:

Turkish firms are classified as SME or large scale firms according to the criteria defined by the Ministry of Industry and Technology. For a firm to be defined as SME, the number of employees should be less than 250 and yearly net sales revenue or total assets should be less than 125 million TL. Firms not included in this definition are called large-scale firms. The asset size of the corporate sector in Turkey is double the GDP, and about half of it belongs to large scale firms. Thus, credit risk evaluation and loan default risk evaluation became very important to financial institutions that provide loans to businesses.
The in-depth credit risk analysis in banks has a very important impact on the quantity of Nonperforming loan detection’s (NPL). Predicting NPL’s before they become non-performing is important for banks since the consequences are excruciating unless provisions are made and for this purpose, early data analytics techniques are used. NPL predictions are aimed to determine the likelihood of a default on credit obligations by a corporation or sovereign entity over a particular time horizon.
The motivation is to predict probable NPL’s before they become non-performing to provide institutions enough time to make provisions. This challenge chooses to focus on large scale firms loans since credit risk accumulation and cyclicality for small and medium size enterprises (SME) and large scale companies may differ, default probabilities for these companies is also examined separately. YKT aims to identify the large scale companies that are probable to be default through analysing their financial data. Combining these with the state-of-the-art machine learning methods and experience in the field, YKT intends to predict non-performing loans before 6 months of period and decrease the ratio of NPL.

Expected outcomes:

This challenge is designed to employ Machine Learning (ML) algorithms on financial data. Especially, feature engineering techniques on time series tabular data is focused on during the project development.

How do we apply?

Read the Guidelines for Applicants

Doubts or questions? Read more about REACH on the About Us page,

have a look at our FAQ section or drop us an email at opencall@reach-incubator.eu.