Abstract: This tutorial will cover an overview of different areas of using NLP in Ecommerce. Specifically we will drill down to sentiment analysis of reviews and attribute extraction. We can cover a brief introduction to different types of sentiment analysis. We will delve deep into a ‘Amazon Reviews’ dataset. We will see how we can solve it using unsupervised and supervised techniques. We will also cover key techniques of attribute extraction. All examples are taken from my book Practical Natural Language Processing (https://www.amazon.in/Practical-Natural-Language-Processing-Python/dp/148426245X)
Session Outline
Part 1: Intro to NLP in Ecommerce:
We look at the classic use cases of NLP in Ecommerce - Product name matching for competitive pricing, Customer Service, Search and Review mining. We will deep dive into review mining
Part 2: Sentiment Analysis
We will see different dimensions of sentiment analysis followed by usecases of sentiment analysis in other industries. We also see possible pitfalls in sentiment analysis
Part 3: Unsupervised Sentiment Analysis
We take the Amazon dataset and start with the most basic form of sentiment analysis. We then look at the errors and correct our unsupervised heuristic approach. We look at VADER library here - a commonly used approach We also start to measure our performance of the unsupervised approach
Part 4: Supervised ML model in Sentiment Analysis
We continue with our unsupervised approach and convert this into a machine learning model. We compare the ML model with the output we got from the unsupervised approach and identify next steps
Background Knowledge
Python, Basic Machine Learning
What is needed to run the codes
Collab notebooks, Sklearn packages, NLTK, Vader, Tensorflow, Keras.
Bio: Mathangi is a renowned data science leader in India. She has 11 Patent grants and 20+ patents published in the area of intuitive customer experience, indoor positioning and user profiles. She has recently published a book - “Practical Natural Language Processing with Python” She has 17+ years of proven track record in building world-class data sciences solutions and products. She is adept in machine learning, text mining, NLP technologies & tools. She is currently heading the data organization of GoFood, Gojek. In the past, she has built data sciences teams across large organizations like Citibank, HSBC, GE, and tech startups like 247.ai, PhonePe. She advises start-ups, enterprises, and venture capitalists on Data Science strategy and roadmap. She is an active contributor on machine learning to many premier institutes in India. She is recognized as one of “The Phenomenal SHE” by Indian National Bar Association in 2019.
