a. Description about the ICP

In this ICP we have 3 tasks

1. Run sentiment analysis on a sample sentence
2. Use GridSearch to find the best accuracy from epoch and batch size options
3. Differentiate between real and spam messages

b. Objective

Our objective for this ICP is to create a neural network that is capable of accurately analyzing words for sentiment related to texts.

c. Design/ implementation(include screenshots with explanations)

Save the model from the source provided

Load model and analyze sample text

Output

Options of Batch size and Epochs to test the best results from the combination

Output

Only a few changes required from original source, such as the result being binary now instead of 3

Output

d. Video link

https://youtu.be/OioznRMrdQs

e. Conclusion

i. The accuracy on the analysis of sentences was more difficult due to words being more complex and context reliant.

ii. What the challenge you have faced

Like previous times, the analysis can be resource hungry and can take quite a lot of time at execution.