Custom Named Entity Recognition for Resumes Using Azure Services

Question

I am working on a project to create a custom named entity recognition (NER) model for extracting information from resumes. i am totally new to this . I would like to know:

1. Which service is better suited for this task, Azure Language Studio or Azure Machine Learning Studio?
2. What type of dataset should I use for training the NER model is there any sample dataset available to train azure ml model?
3. Which pre-trained model is recommended for fine-tuning for resume data?
4. Is there a recommended workflow or best practices for achieving this using Azure resources?

Any guidance or resources would be greatly appreciated.

Answer 1

Hello @Dinnemidi Ananda Kumar , Thanks for using Microsoft Q&A Platform.

I believe natural language processing tasks like NER may be a better fit for your project. Specially, creating a custom NER project where you can train custom models for custom entity recognition to do a certain task. These models extract named entities and are trained by learning from tagged data.

I will suggest you to explore this model if that fits your use case: https://learn.microsoft.com/en-us/azure/ai-services/language-service/custom-named-entity-recognition/quickstart?pivots=language-studio

I hope this helps.

Regards,

Vasavi