Hello Community, I'm working with a large number of scanned exams from various clients, which I need to process using Azure AI Document Intelligence (Form Recognizer). My goal is to extract the clients' first names, last names, and middle names, then update clients' score. The process is as follows: If the extracted names match a record in Database, the system will retrieve the score and update it in the client's exam record in the database. In cases where the names match more than one record, "MyApp" then compare the client's address for a more accurate match. If there's no match or an error occurs, "MyApp" will send the exam to a staging container where it awaits manual review by an agent. However, I'm facing some challenges with this logic in "MyApp." The clients' names are in English but follow varied naming conventions from different countries, notably Indian and Chinese styles. Moreover, the names aren't always in a consistent order or format. For example, I encounter formats like: First name + Last name Last name, First name Is there a way to accurately determine which part of the name is the first name to improve pattern matching in these scenarios?

FullName in Document Intelligence

Ming 30

Hello Community,

I'm working with a large number of scanned exams from various clients, which I need to process using Azure AI Document Intelligence (Form Recognizer). My goal is to extract the clients' first names, last names, and middle names, then update clients' score.

The process is as follows:

If the extracted names match a record in Database, the system will retrieve the score and update it in the client's exam record in the database.
In cases where the names match more than one record, "MyApp" then compare the client's address for a more accurate match. If there's no match or an error occurs, "MyApp" will send the exam to a staging container where it awaits manual review by an agent.

However, I'm facing some challenges with this logic in "MyApp." The clients' names are in English but follow varied naming conventions from different countries, notably Indian and Chinese styles. Moreover, the names aren't always in a consistent order or format. For example, I encounter formats like:

First name + Last name
Last name, First name

Is there a way to accurately determine which part of the name is the first name to improve pattern matching in these scenarios?

santoshkc 6,955 Reputation points Microsoft Vendor

2023-11-16T10:27:39.58+00:00

Hi @Ming ,

Thank you for reaching out to Microsoft Q&A forum!

Regarding the order of names, it's not always guaranteed that the first name comes before the last name, especially when dealing with names from different cultures.

You should have redesign your database better so that the data will be inserted from the start to three explicit columns for first name, last name, and middle name, this can be a good solution if the data is always entered in a consistent format. Any solution we will provide will never give you the guarantee that the result is as expected if you cannot control the input in this case!

I hope you understand. Thank you!
santoshkc 6,955 Reputation points Microsoft Vendor

2023-11-17T07:57:14.5033333+00:00

Hi @Ming ,

We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. In case if you have any resolution, please do share that same with the community as it can be helpful to others. Thank you.
santoshkc 6,955 Reputation points Microsoft Vendor

2023-11-19T03:51:21.3366667+00:00

Hi @Ming ,

Following up to see if the given information was helpful. In case if you have any resolution, please do share that same with the community as it can be helpful to others. Thank you.
Ming 30 Reputation points

2023-11-19T23:49:59.6966667+00:00

Hi Santoshkc,

Thank you for your answer. I think this is the only way to do this at this point. I really appreciate for your help and suggestion. Thanks!
Deleted

This comment has been deleted due to a violation of our Code of Conduct. The comment was manually reported or identified through automated detection before action was taken. Please refer to our Code of Conduct for more information.

Share via

FullName in Document Intelligence