Subscribe to Our Newsletter

Success! Now Check Your Email

To complete Subscribe, click the confirmation link in your inbox. If it doesn’t arrive within 3 minutes, check your spam folder.

Ok, Thanks
South Korean Fintech BalanceHero Expands AI Capabilities in India
Source: BalanceHero

South Korean Fintech BalanceHero Expands AI Capabilities in India

Korean fintech BalanceHero upgrades AI-powered credit scoring system to improve loan approvals and risk assessment for India's underserved consumers.

Philip Lee profile image
by Philip Lee

Seoul, South Korea - BalanceHero announced on Tuesday that it has expanded artificial intelligence capabilities in its Alternative Credit Scoring System (ACS) for the Indian market.

The system upgrade incorporates large language models (LLM) to increase loan approval rates for India's middle and low-credit consumers while adding a new default prediction model capable of assessing credit risk through SMS text analysis.

The company has implemented an AI Named Entity Recognition (NER2) model, expanding its institutional data coverage to include regional small banks. 

The upgrade has increased bank coverage by sixfold and lending company coverage by threefold compared to the previous system.

The company said the enhanced system now processes more than double the previous data volume, enabling more detailed extraction of account names, bank names, and deposit amounts.

BalanceHero's ACS system, first developed in 2019, has historically increased financial product approval rates by up to 140% while reducing default rates by more than 10% annually.

The upgrade incorporates a novel automated AI assessment generation feature that furnishes prompt feedback on product approvals and rejections, enhancing customer response times.

Moreover, the company has developed a novel Large Language Model (LLM) for default prediction.

The model could quantify the default probability based exclusively on customers' financial activity data.

This facilitates more accurate risk management.

BalanceHero, which serves India's population of 1.4 billion, built its machine learning-based alternative credit scoring system to provide digital financial services to middle and low-credit consumers who previously lacked credit history.

A company representative said the continuous upgrade of machine learning and AI technology would help extend financial services to more Indian consumers with limited credit access. 

They stated their commitment to expanding financial service accessibility in the Indian market using Korean technology and systems.

Philip Lee profile image
by Philip Lee

Subscribe to The Pickool

Success! Now Check Your Email

To complete Subscribe, click the confirmation link in your inbox. If it doesn’t arrive within 3 minutes, check your spam folder.

Ok, Thanks

Read More