Analyzing Seed Variety

Lab 3.1

Luke Stucky

Viewing the Data

[1] "Top 10 variables in the Dataset:"

   compactness width of kernel variety
1       0.8710           3.312    Kama
2       0.8811           3.333    Kama
3       0.9050           3.337    Kama
4       0.8955           3.379    Kama
5       0.9034           3.562    Kama
6       0.8951           3.312    Kama
7       0.8799           3.259    Kama
8       0.8911           3.302    Kama
9       0.8747           3.465    Kama
10      0.8880           3.505    Kama

[1] "Bottom 10 Variables in the Dataset:"

    compactness width of kernel  variety
201      0.8609           2.989 Canadian
202      0.8977           3.135 Canadian
203      0.8680           2.810 Canadian
204      0.8874           3.091 Canadian
205      0.8567           2.960 Canadian
206      0.8783           2.981 Canadian
207      0.8511           2.795 Canadian
208      0.8883           3.232 Canadian
209      0.8521           2.836 Canadian
210      0.8684           2.974 Canadian

Viewing the top 10 and bottom 10 variables already shows a clear distinction between the Canadian and Kama kernels.

Summary of Data

[1] "Summary of the Kernel Compactness:"

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
 0.8081  0.8569  0.8734  0.8710  0.8878  0.9183

[1] "Summary of the Kernel Width:"

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  2.630   2.944   3.237   3.259   3.562   4.033

[1] "Table of the Kernel Variety:"


    Kama     Rosa Canadian 
      70       70       70

Each seed variety has 70 observations which entails a balanced distribution.

Visualization of Seed Data

Graph Findings:

The Canadian kernel, on average, has the smallest width as well as the smallest compactness. The Rosa kernel has the largest width along with a larger compactness. The Kama kernel has a width that is in between the Rosa and the Canadian kernels while having a compactness similar to the Rosa kernel. Each variety is distinct from the others and can be separated into clusters fairly easily.