Sentiment Analysis of Twitter Data (Using CNNs)
- 0 Collaborators
Sentiment Analysis of Twitter Data using a CNN with character-level encoding in Keras and Tensorflow. ...learn more
Project status: Under Development
Groups
DeepLearning,
Artificial Intelligence India,
Student Developers for AI
Overview / Usage
A deep learning model to implement Sentiment Analysis by a character-level Convolutional Neural Network, based on the paper by Xiang Zhang, Yann LeCun, and Junbo Zhao. Model is scaled down to decrease training time and memory usage.
Methodology / Approach
This model has been implemented using Keras(Tensorflow backend), Pandas and Numpy. Model Architecture:
1D Conv Layer with 64 kernels, kernel_size=7, ReLU activation.
1D Max Pooling with pool_size=3
1D Conv Layer with 64 kernels, kernel_size=7,ReLU activation.
1D Max Pooling with pool_size=3
1D Conv Layer with 32 kernels, kernel_size=3,ReLU activation
1D Conv Layer with 32 kernels, kernel_size=3,ReLU activation
1D Conv Layer with 16 kernels, kernel_size=3,ReLU activation
1D Conv Layer with 16 kernels, kernel_size=3,ReLU activation
1D Max Pooling with pool_size=3
Flattening
Dense Layer with 64 nodes, ReLU activation, 0.5 dropout
Dense Layer with 32 nodes, ReLU activation, 0.5 dropout
Output Node with sigmoid activation
Technologies Used
Python, Numpy, Pandas, Tensorflow, Keras