This project is designed to practice the perceptual mapping and multidimensional (MDS) plot learned in the Marketing Analytics course using R programming language. The main goal is to pick a product category and analyze the brands within that category and see if there are any insights to generate.
- Pick a brand based on personal preference/research need
- Calculate the metric - cooccurrence
- Calculate the metric - lift
- Based on lift, build a dis-similarity matrix
- Build MDS plot
- There are different ways to calculate cooccurrence and lift; tread carefully
- transform similarity matrix based on cooccurrence and lift to dissimilarity matrix because the MDS is based on dissimilarity
- Interpretation of the MDS plot
https://nijianmo.github.io/amazon/index.html
Justifying recommendations using distantly-labeled reviews and fined-grained aspects
Jianmo Ni, Jiacheng Li, Julian McAuley
Empirical Methods in Natural Language Processing (EMNLP), 2019