• Leveraging pre-trained models, including PhoBERT and a Vietnamese accent marker model to handle the challenges of processing Vietnamese text, determine whether comment is positive, negative or neutral, ultimately enhancing customer satisfaction.

  Link of PhoBERT model: https://pdfs.semanticscholar.org/74fc/832dd6c77253595cf3c1c852045c8da93c13.pdf

  Link of Transformer model for inserting Vietnamese accent marks: https://huggingface.co/peterhung/vietnamese-accent-marker-xlm-roberta

• Extracting data from MongoDB, cleaning and transforming data using python.

  

  

Code script for sentiment analysis Vietnamese (with and without accent). Implementing efficient batch processing techniques to handle large-scale data.

• Building Dashboard using Power BI and providing feasible insights for improving user experience.