Origin of wine part 1
Introduction
I have written 13 blogs about how I did Wine Review task from Kaggle. You can find it here part1. It was done by using winemag-data-130k-v2.csv dataset. The model is learned to predict variety which used to produce wine.
In this series, I am going to predict possible location of wine by wine’s description.
Imaging I can taste a glass of wine and write down description then machine can suggest me a list of possible of wine similar to my description. Furthermore it can provide the place where each wine came from.
Task
The task has already been defined from Kaggle.
My overall goal is to create a model that can identify the variety, winery, and location of a wine based on a description.
The objective is to identify wines and its associated information such as variety, winery and location base on wine’s description that is given by user.
Framework
I choose Scikit-Learn for machine learning framework to build the model.
Dataset
There are 2 datasets in Kaggle both winemag-data-130k-v2.csv and winemag-data_first150k.csv respectively. I am using winemag-data_first150k.csv as dataset to train the model.