AmazonHackOn Jatin104_70b4

Augmented Reality:

An AR/VR based regular item-at-your-place check using the backcamera of your phone, to help customers be more decisive and reduce return rate To Run:

Start index.html

Authenticate Products:

Using a blockchain based Crypto tag to remove counterfeit products, improve tracking of products in supply chain, authentication and validation Crypto tags will help in detecting counterfeit products, tracking product as it moves through the supply chain and enable any NFC enabled device to provide authentication and validation details of the product. This will ensure a secure and authenticated blockchain database on an effectively low budget.

Reliability Score Prediction:

Our focus is to enhance the reliability on product. With help of sentiment analysis and anomaly detection, we will find the percentage of positive reviews of that particular product Combining these two factors, we will take out the reliability score of the product. Technologies used will be sentiment analysis model and anomaly detection.

Fake reviews removal: Naive bayes model was used, on the amazon reviews dataset with accuracy of 80%.

1.2 Sentiment Analysis: Support vector classifier model was used with accuracy of 94.5%

1.3 Predicts the score by combining the both models.