Using technologies of your choice, develop a simple program to rank-order Walmart product recommendations based upon customer reviews.
The program should invoke the Walmart Labs Open API to implement the following workflow:
- Search for products based upon a user-provided search string
- Use the first item in the search response as input for a product recommendation search
- Retrieve reviews of the first 10 product recommendations
- Rank order the recommended products based upon the review sentiments
$git clone https://github.com/ZhengC1/CodeChallenge.git
or
download the zip
The Sentiment Analysis will take about 20 to 30 seconds
$python CodeChallenge.py
For this project, i would have preferred even a mock json file of customers and
their purchases so that i could see to what percentage my recommendations aligned
with their product purchases
The Test.py file is essentially the CodeChallenge.py file but much wordier and prints out multiple reviews
$python Test.py
Python 2.7 +
Git
Uses PEP8 for python Style
- Walmart Open API
- Natural Language Sentiment Analysis API
- Ubuntu 16.04
- Gnome-Terminal
- VIM - Visual Improved
* assuming that "ordering the products based on review sentiment means sentiment analysis of reviews
* Using the first 5 reviews for each product
- Chun Zheng -
- Other projects - ZhengC1
- Manual Implementation of Sentiment Analysis with Twitters API and NODEjs - AppHack7
- Thank you to all the programmers who made this possible - I.E. ME :D
- Thank you Walmart for making an API