AlexaSentimentAnalysis

Trained a Deep Neural Network to perform Natural Language Proceessing and analyze whether a customer’s review on Amazon’s product Alexa was positive or negative with a 92% accuracy on test data. Dataset consists of 3000 Amazon customer reviews, star ratings, date of review, variant and feedback of various amazon Alexa products. The objective is to discover insights into consumer reviews and perfrom sentiment analysis on the data.

Libraries used:

Data preprocessing and visualization:

• sklearn, pandas, numpy, matplotlib and seaborn

Model Training and Evaluation:

• Tensorflow with Keras and sklearn

Acknowledgements:

• Tensorflow 2.0 Practical: https://www.udemy.com/course/tensorflow-2-practical

• https://www.kaggle.com/c/acc-german-traffic-sign-classification/data