AI/ML for a loan application
Budget: $750 – $1,500 USD
I am looking for an artificial intelligence (AI) that can be integrated to a newly built loan app and probably for other loan operations. The AI will help to improve KYC in order to prevent/ reduce loan defaults.
POSSIBLE CAPABILITIES OF THE AI
The AI can use some/all of the following information from the following proposed sources for KYC.
Geolocation data: The AI determines the geographical location of person by means of digital information. Call location, provides insights about subscribers’ income level, job stability, job rotation, and so on. If s/he calls from a residential area, their risk profile will be more credit favorable
Social data: The AI uses data from social media platforms like Facebook, LinkedIn, Yahoo, Google (including emails) and Twitter to map relationships with 120 social media profiles. The predictive power of social network data is based on the principle of ‘who you know matters’. In other words, borrowers’ personality and relationships are pivotal to assess risk.
Smartphone data: The AI can collect telco data for risk profiling. This AI uses call/SMS metadata (when, where and duration), top-up data and value-added- service transaction data to determine borrowers’ income, mobility patterns, financial skills, consumption profile, social capital, and life habits.
Digital profiles
Transactional data: The AI uses bank transaction history, e-commerce transaction data, Utility bills, Spending, download of financial apps and earning habits etc.
I am open to any idea on how/suggestions/design on how any AI can work
POSSIBLE CAPABILITIES OF THE AI
The AI can use some/all of the following information from the following proposed sources for KYC.
Geolocation data: The AI determines the geographical location of person by means of digital information. Call location, provides insights about subscribers’ income level, job stability, job rotation, and so on. If s/he calls from a residential area, their risk profile will be more credit favorable
Social data: The AI uses data from social media platforms like Facebook, LinkedIn, Yahoo, Google (including emails) and Twitter to map relationships with 120 social media profiles. The predictive power of social network data is based on the principle of ‘who you know matters’. In other words, borrowers’ personality and relationships are pivotal to assess risk.
Smartphone data: The AI can collect telco data for risk profiling. This AI uses call/SMS metadata (when, where and duration), top-up data and value-added- service transaction data to determine borrowers’ income, mobility patterns, financial skills, consumption profile, social capital, and life habits.
Digital profiles
Transactional data: The AI uses bank transaction history, e-commerce transaction data, Utility bills, Spending, download of financial apps and earning habits etc.
I am open to any idea on how/suggestions/design on how any AI can work