Using Fast.AI with Transfer Learning to Build Fine-grained Sentiment Analysis


Conventional sentiment analysis involves providing a binary classification of the sentiments of text data (just positive or negative). Yet, that is lacking since the subtleties and intensities in human language can hardly be captured with a naïve binary sentiment model.

Yet, creating a fine-grained sentiment model with granular classification of sentiment can be challenging with conventional rule-based models such as Vader. Fortunately, the introduction of transfer learning to NLP has paved the way for us to build high quality multiclass sentiment models. In this training, we will showcase how to apply transfer learning to create fine-grained sentiment analysis with the fastai and transformers library.

Session Outline
Part 1: Differences between binary and fine-grained sentiment analysis
You will learn about the differences between a binary and multiclass sentiment model, and appreciate the challenges of building a multiclass sentiment model

Part 2: Transfer learning
You will learn about the broad concept of transfer learning, and appreciate why the introduction of transfer learning to NLP is a huge game changer

Part 3: Build your own fine-grained sentiment model with transfer learning
This part of the workshop will be a code-along session where you will build your own fine-grained sentiment model using transfer learning with the fastai and transformers library

Background Knowledge
- Python programming
- Machine learning models
- Jupyter Notebook


David serves as an advisor to the Data Science curriculum team at Heicoders Academy, a fast-growing tech education training provider based in Singapore. Before that, he co-founded another renowned tech education company in Singapore, Hackwagon Academy.
Previously a Machine Learning Engineer at Droice Labs, a New-York based AI company in the healthcare sector, David has multiple technical consulting experiences under his belt, including a 1-year stint with Louis Vuitton in the US. He has a Master of Management Science & Engineering from Columbia University, and a Bachelor of Information Systems from Singapore Management University, where he graduated as a valedictorian.
In his free time, fueled by his passion to democratise data science, David contributes articles to Medium on topics like workplace automation and machine learning techniques.

Open Data Science




Open Data Science
One Broadway
Cambridge, MA 02142

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from Youtube
Consent to display content from Vimeo
Google Maps
Consent to display content from Google