Mall Customers Segmentation Using K-Means Clustering Analysis
Explore the results of K-means clustering analysis on mall customers, revealing insightful patterns in their spending behavior based on income levels. Discover the distinct customer clusters and their unique spending habits in this detailed dataset analysis presented by Dr. Panagiotis Repoussis.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
Clustering Mall Customers Dr. Panagiotis Repoussis
MALL CUSTOMERS SEGMENTATION Dataset view
MALL CUSTOMERS SEGMENTATION Descriptive Statistics
MALL CUSTOMERS SEGMENTATION Descriptive Statistics
Preprocessing (features selection and exclusion)
We dont need to include demographics (gender and age), but we can gain valuable insight by referencing them at our analysis. Preprocessing
Putting data in SPSS Dataset Preview Attribute Description 200 Customers Both Males and Females Ages ranges from 18 to 70 Annual Income varies from 15K to 137K Spending Score from 1 to 99
Putting data in SPSS K-means Clustering using attributes: AnnualIncome$ SpendingScore1100 K-Means Algorithm
Putting data in SPSS Variables: AnnualIncomek$ SpendingScore1100 Method: Iterate and Classify Number of Clusters: 4 5
Results of k-means clustering in SPSS Results: Initial interpretation of the clusters: Cluster 1: Customers with an average Income spending at an average pace. Cluster 2: Customers with high income but spending less than expected Cluster 3: Customers with low income spending less Cluster 4: Customers with high income spending accordingly Cluster 5: Customers with low income spending unconsciously more than they can afford.
MALL CUSTOMERS SEGMENTATION CLUSTERING K-MEANS
Hierarchical clustering We also try other methods such as Hierarchical clustering using Ward linkage. Results are very similar. Differences with KMeans
MALL CUSTOMERS SEGMENTATION CLUSTERING K-MEANS HIERARCHICAL CLUSTERING Cluster 0 , Size: 22 , Center: [25.72 79.36] Cluster 0 , Size: 32 Cluster 1 , Size: 81 , Center: [55.29 49.51] Cluster 1 , Size: 85 Cluster 2 , Size: 39 Cluster 2 , Size: 35 , Center: [88.2 17.11] Cluster 3 , Size: 21 Cluster 3 , Size: 39 , Center: [86.53 82.12] Cluster 4 , Size: 23 Cluster 4 , Size: 23 , Center: [26.30 20.91] Silhouette score: 0.55299 Silhouette score: 0.55393
MALL CUSTOMERS SEGMENTATION Descriptive Statistics per cluster Cluster 0 Mean Age: 25.27 Annual Income Mean: 25.73 min: 15 max: 39 Spending Score Mean: 79.36min: 61 max: 99 Size: 22 Cluster 3 Mean Age: 32.69 Annual Income Mean: 86.54 min: 69 max: 137 Spending Score Mean: 82.13 min: 63 max: 97 Size: 39 Cluster 4 Mean Age: 45.22 Annual Income Mean: 26.3 min: 15 max: 39 Spending Score Mean: 20.91 min: 3 max: 40 Size: 23 Cluster 1 Mean Age: 42.72 Annual Income Mean: 55.3 min: 39 max: 76 Spending Score Mean: 49.52 min: 34 max: 61 Size: 81 Cluster 2 Mean Age: 41.11 Annual Income Mean: 88.2 min: 70 max: 137 Spending Score Mean: 17.11 min: 1 max: 39 Size: 35
MALL CUSTOMERS SEGMENTATION Descriptive Statistics per cluster
MALL CUSTOMERS SEGMENTATION Descriptive Statistics per cluster Mean Values of each Cluster 88.2 86.54 90 82.13 79.36 80 70 55.3 60 49.52 45.22 50 42.72 41.11 40 32.69 26.3 25.73 30 25.27 20.91 17.11 20 10 0 Cluster 0 Cluster 1 Cluster 2 Cluster 3 Cluster 4 AGE ANNUAL INCOME SPENDING SCORE
MALL CUSTOMERS SEGMENTATION K-means clustering interpretation Most customers are women in every case Young customers (mean age <= 35) have the highest spending score Cluster 0 - high spending score, very young women even with low annual income (mass consumers) Cluster 1 - middle aged (mean age >= 35) customers with balanced income and spending score (mainstream) Cluster 2 - middle aged men with high annual income but low spending Score (careful buyers) Cluster 3 - young women with high income and high spending score (big spenders) Cluster 4 - low income, low spending older women (hesitant - conservatives)
Two step cluster analysis with SPSS Silhouette score: 0,7
