https://github.com/shridhar1504/wine-customer-segment-classification-using-lda-datascience-project
This project is to classify the Customers who prefers three types of wine and by using Linear Discriminant Analysis(LDA) after reducing the dimensions we can able to classify the Customers into 3 segments based on their preference.
https://github.com/shridhar1504/wine-customer-segment-classification-using-lda-datascience-project
classification-algorithm classification-models data-science dimensionality-reduction linear-discriminant-analysis machine-learning machine-learning-algorithms supervised-learning
Last synced: about 1 year ago
JSON representation
This project is to classify the Customers who prefers three types of wine and by using Linear Discriminant Analysis(LDA) after reducing the dimensions we can able to classify the Customers into 3 segments based on their preference.
- Host: GitHub
- URL: https://github.com/shridhar1504/wine-customer-segment-classification-using-lda-datascience-project
- Owner: shridhar1504
- Created: 2023-06-26T07:01:42.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2023-07-02T07:33:02.000Z (almost 3 years ago)
- Last Synced: 2025-02-15T06:29:01.369Z (over 1 year ago)
- Topics: classification-algorithm, classification-models, data-science, dimensionality-reduction, linear-discriminant-analysis, machine-learning, machine-learning-algorithms, supervised-learning
- Language: Jupyter Notebook
- Homepage:
- Size: 722 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Wine-Customer-Segment-Classification-using-LDA-Datascience-Project
This project is to classify the Customers who prefers three types of wine and by using Linear Discriminant Analysis(LDA) after reducing the dimensions we can able to classify the Customers into 3 segments based on their preference.
## Problem Statement:
The wine industry is a complex and competitive market. There are thousands of different wines available, and it can be difficult for customers to know which wines they will enjoy. One way to help customers make better wine choices is to develop a wine classification system that can be used to categorize wines based on their flavor profiles.
## Solution Approach:
One way to develop a wine classification system is to use machine learning algorithms to analyze wine sensory data. This data can include information such as the color, aroma, flavor, and aftertaste of wine. By analyzing this data, machine learning algorithms can identify patterns that can be used to classify wines into different categories which customers prefer.
## Observations:
The project found that the following factors are most important in prediction of classification of customers based on their wine tastes:
* Alchol Content
* Malic Acid Content
* Ash Content
* Ash Alcanity Content
* Magnesium Content
* Total Phenols Contents
* Flavanoids Contents
* Non - Flavanoids Contents
* Proanthocyanins Contents
* Color Intensity of Wine
* hue of the Wine
* OD280(Protein Concentration Determination of various Wines)
* Proline
## Findings:
It is possible to use machine learning algorithms to classify wines into different categories which customers prefers with a high degree of accuracy. The study found that the most important factors for classifying wines were the color, aroma, and flavor of the wine.
## Insights:
The findings of this study have several implications for the wine industry. First, it suggests that machine learning algorithms can be used to develop accurate wine classification systems which customers prefer. This could help wineries make better marketing based on the customer's wine choices and could also help customers to choose their wines more effectively. Second, the study suggests that there are significant differences in the flavor profiles of different wine varieties. This means that wineries can use the results of this study to find wines that they are more likely to enjoy by the customers.
## Conclusion :
Achieved in developing a predictive model to predict the preference of the wine by Customer's choice with accuracy of 100% for several models such as Logistic Regression, Random Forest Classifier, Extra Trees Classifier and CatBoost Classifier.