We shall train three separate Neural Networks, namely: a Simple Neural Net, a Convolutional Neural Net and a Long Short Term Memory Neural Net. LSTM networks are actually considered to be quite suitable for handling NLP problems, and by the end of this video, you will understand why.
Sentiment Classification LSTM Model we shall be training as part of this tutorial is so good that it not only predicts the user movie review sentiment as positive/negative, it also does a fabulous job in predicting IMDb Rating itself corresponding to these reviews with mind blowing accuracy.
🔥 During the course of next ~25mins, we shall discuss:
a. Word Embedding (which is a powerful feature extraction technique),
b. Intuition on various Neural Network approaches (Simple / CNN /LSTM),
c. Concept of Overfitting (that happens when a model memorises training data, rather than understanding general distribution)
d. And, of course, our high-level solution architecture for training our Sentiment Classification Models
🔥 Sections:
00:00 Introduction
02:17 Solution Architecture
03:50 Word Embeddings, Neural Networks & Overfitting
08:56 Free Skillcate Project Toolkit
09:55 Code: Intro
11:42 Code: Data Preprocessing
13:44 Code: Tokenizer & Embedding Layer
15:50 Code: Simple NN Training
17:50 Code: CNN Training
19:27 Code: LSTM Training
21:03 Having fun with Model
22:59 Updates on Free 1:1 Mentoring
🔥 Important Links:
Toolkit link: https://drive.google.com/drive/folders/1TK9k41RT8Nf3IhzerNWHpEqWztsk2gAP?usp=sharing
Sentiment Analysis Project based Review Classification Project from Skillcate: https://youtu.be/zwR6M5zpnWs
Sentiment Analysis Project (End-to-end) with ML Model Building + Deployment (using Flask):
- Model Building: https://youtu.be/lKAdxN0qrgk (Part-1)
- Model Deployment: https://youtu.be/KEQCVwJU5KU (Part-2)
GitHub Repo Link: https://github.com/skillcate/movie-sentiment-analysis-with-deep-neural-networks.git
🔥 Do like, share & subscribe to our channel. Keep in touch:
Email: [email protected]
Website: https://www.skillcate.com
Facebook: https://www.facebook.com/groups/mlprojects
Telegram: https://t.me/skillcate