Deep Learning with Python and Keras: Build a Model For Sentiment Analysis (2024)
1h 56mAdvanced2024-02-21
Authors
Janani Ravi
Certified Google Cloud Architect and Data Engineer
Course details
Learn to apply sentiment analysis to your problems through a practical, real world use case. In this course, certified Google cloud architect and data engineer Janani Ravi guides you through the process of building and training a RNN to do sentiment analysis, including validating your results. Go over how to preprocess text for sentiment analysis, as well as approaches you can use and challenges you may encounter. Get set up with Google Colab and import Python modules and loading data, then learn how to analyze word lengths, clean and preprocess text, and visualize text with word clouds. Explore feed-forward neural networks, then dive into configuring, training, and evaluating your dense neural network (DNN). Plus, learn how to train RNNs and LSTNs.
Skills covered
KerasMachine LearningAdvancedPythonArtificial Intelligence (AI)Open Source
Concepts
Introduction
- An overview of sentiment analysis
- Prerequisites
Overview of Sentiment Analysis
- Preprocessing text for sentiment analysis
- Word vector encodings and word embeddings
- Types of sentiment analysis
- Approaches and challenges in sentiment analysis
Cleaning and Preprocessing Text Data
- Getting set up with Google Colab
- Importing Python modules and loading data
- Analyzing word lengths across sentiment categories
- Cleaning and preprocessing text
- Visualizing text using word clouds
Sentiment Analysis Using Dense Neural Networks
- Feed-forward neural networks
- Splitting data into training test and validation sets
- Representing text using count vectorization
- Configuring the dense neural network (DNN)
- Training and evaluating the DNN
- Configuring the count vectorizer as a model layer
- Representing text using TFIDF vectorization
- Training and evaluating the model
- Representing text using integer sequences
- Training ADNN using embeddings
Sentiment Analysis Using Recurrent Neural Networks
- Recurrent neural networks
- Long memory cells
- The LSTM and GRU cells
- Training a recurrent neural network
- Training an LSTM network
- Serializing a model to disk and loading the model
Conclusion
- Summary and next steps