Predicting Subrogation
Get ahead of the claim assignment.
Making a decision regarding the subrogation potential of a claim requires reviewing all facts of the claim. Relevant facts can be dispersed through numerous free text notes added at random times during the lifetime of the claim. Claim handlers are overloaded with the sheer volume of claims that require timely processing: they have no time to investigate all past records for facts potentially important for subrogating each time a new claim note is loaded. This is why potential subrogation opportunities are frequently overlooked resulting in loss of revenue.
Solution
Megaputer offers a solution, called SubroPredict, for timely and reliable prediction of subrogation opportunities based on the automated analysis of claim notes. For every claim it extracts all relevant facts out of a collection of several hundred facts that might be relevant for making the subrogation decision. Then SubroPredict sends the extracted facts through an artificial intelligence (AI) model predicting the probability and potential amount of subrogation based on all available data points.
How it works
SubroPredict performs sophisticated text analysis and fact extraction based on a combination of advanced linguistic and semantic analysis, along with pattern recognition and entity resolution techniques. It provides unparalleled depth and accuracy for identifying people, their roles, vehicles, actions, and other facts that enable the correct interpretation of information to be recorded as free text.
Every time a new note is loaded, the solution, Megaputer’s PolyAnalyst software, extracts relevant facts, adds them to the list of already stored data points for the considered claim, and then applies the AI model to predict the subrogation probability of the claim. This facilitates early detection of subrogation opportunities (as well as subrogation threats).
To identify potential subrogation opportunities, a large P&C insurance company was relying on a team of analysts who manually review claim notes. This process resulted in a high percentage of missed subrogation. The company realized its need for an automated solution to extract relevant information from claim notes and use it to identify good subrogation opportunities across all claims accurately and sufficiently early.
Benefits
