A carregar...
Publicação
Claims processing automation - Modernization of an insurance company internal process
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2.44 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Deep learning and text mining are involved in the research. This work includes the project I developed
together with my colleagues at SAS Institute during my internship experience. In this project we had
to support an Insurance company for the automation of their existing claim processing system. In fact,
as of today, the procedure of reading the incoming claim requests, selecting the useful information
and extracting it to a data management software, is done manually for hundreds of claims every day.
The job required by the insurance company is to substitute the existing procedure with an automated
one, by implementing an OCR system to read the raw data contained in the documents sent by the
customers and transform it into clean and useful information to be inserted into the data management
software.
This research will show the investigation on how to deal with this problem and the objective is to
automate the classification of the documents for the company, to provide them a system to prioritize
the most urgent documents and to execute some technical and administrative checks on the extracted
information. The automation is shown to be feasible; the completeness and accuracy of the
information extracted are solid, proving that this specific task in the insurance company sector can be
realized and help to reduce costs while improving time performance.
Descrição
Internship Report presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Data Science
Palavras-chave
Claim processing Automation Optical Character Recognition SAS Text interpretation