The assessment is to assess your ability to design NLP solutions based on the following case. Amazon is one of the world's largest e-commerce companies, with a vast product range. It receives hundreds of thousands of product reviews daily from its users worldwide. These reviews contain valuable information about the product and customer sentiments towards it. However, with the volume of incoming data, it's nearly impossible for Amazon to manually analyse all the reviews to extract actionable insights. This is where NLP comes in. Your task is to create an NLP solution to help Amazon structure and analyse the review data. Here is an Amazon review dataset collected in the range of May 1996 - Oct 2018, including reviews (ratings, text, helpfulness votes), product metadata (descriptions, category information, price, brand, and image features), and links (also viewed/also bought graphs). You are encouraged to use the smaller per-category dense subsets, which have been reduced to extract the k-core, such that each of the remaining users and items have k reviews each. Dataset link: https://cseweb.ucsd.edu/~jmcauley/datasets/amazon_v2//nQuestion la Based on your task, which is to create an NLP solution to help Amazon structure and analyse the review data, determine a scenario/problem that you are going to analyse and design NLP solution for the identified scenario/problem. (word limit: 500) (20 marks) Question 1b Design and implement relevant NLP analysis flow to address the identified questions, which includes but is not limited to: (i) Dataset preparation; (ii) Vocabulary building; (iii) Sentiment analysis; (iv) Named Entity Recognition; (v) Topic modeling

Fig: 1

Fig: 2