Insurance
❯
Manage claims
❯Detection of fraudulent medical claims
Detection of fraudulent medical claims
For:
TPA, medical insurance companies
Scope:
Build an ML model to classify if a particular claim can be fraudulent.
Goal:
Anticipate Risks
AI solution for car damage classification
For:
Insurance companies, car owners/users
Scope:
Car damage classification for common damage types such as bumper dent, door
dent, glass shatter, head lamp broken, tail lamp broken, scratch and smash.
Goal:
Automate a Business Process
From Weeks Down to Hours: How insurers innovate with AI to shorten claims processing time and improve customer experience
For:
- Claims Management
- Customer Experience.
Goal:
Improved Customer Experience
Detection of fraudulent medical claims
For:
TPA, medical insurance companiesGoal:
Anticipate RisksProblem addressed
Upgrade from an only-human-interpretation to ML-assisted fraud detection.
Scope of use case
Build an ML model to classify if a particular claim can be fraudulent.
Description
The third party administrator (TPA) company has a very
good visualization dashboard to eyeball trends by patient, by
doctor and by condition of the medical claims submitted to
the insurance companies the TPA serves. However, the
identification of anomalies from the visual representation
was still done on a subjective judgment basis. The ML model
was developed to identify anomalies in claims that can have
fraudulent activities by the patient, by the doctor or by both
in collusion.
Machine Learning
AI: Understand