how to set up a verification system that uses the customer’s information and image and then notifies systems administrators when something is wrong.
What it does
they car insurers receive a lot of insurance claims from their customers, and the insurer must ensure that every customer’s car is in fact damaged before paying any money. This is a tedious task but luckily a task we automated.
How we built it
This project is built with ReactJS and the following expert.ai flow:
To start the claim verification ,Expert.Ai controls the creation of entities and lemmas from submitted source doc and extracts claimant name and link to photo of damage URL .
1) User Triggers on data entry in source document field
2) Do a Post JSON to the Expert.ai Auth API (https://developer.expert.ai/oauth2/token)
3) With access token returned, create an analysis of source text with a Post Json to Expert.ai api to return the Main entities / lemmas. (https://nlapi.expert.ai/v2/analyze/standard/en/entities) .
If there are any entities/ lemmas found that expert.ai creates, add them to the claim record from the response results
We Built a workflow with Zapier that uses prediction model .
Connect AI model to a Zapier workflow. Connect Expert.ai to React frontend. Connects React to airtable spreadsheet.
Challenges we ran into
Drafting the right sentences to spin up document analysis in expert Api..
find way insurer must ensure that every customer’s car is in fact damaged before paying any
Accomplishments that we're proud of
Put self in the shoes of a car insurer. Build a natural language Ai workflow, that automates a tedious task . Connect expert.ai to a Zapier workflow that uses a prediction model.
What we learned
how to set up a verification system that uses natural language api and the customer’s information and image and then notifies systems administrators when something is wrong. the uses of natural language api