Wild Mushrooms
Classification – Edible or
Poisonous
Yulin Shen
ECE 539 Presentation
2013 Fall
• Mushroom is a kind of food with
high nutrition, however, it is
sometimes poisonous!
• A classification problem.
• Develop some models for
prediction.
Dataset is from UCI Machine Learning Repository
About 8000 combinations
Convert data into numerical form with an integral scale
It can help to improve distances order for K-NN
4-way cross validation
Get 4 pairs of datasets
K-NN Classifier
Euclidean distance
Exhaustive K
Determined by occurrences
Naïve Bayer Classifier
Each probability of features
Compare positive and negative probabilities
Some features have very high probabilities
Results
1
1
0.9
0.9
0.8
0.8
0.7
0.7
0.6
0.6
0.5
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0
0
1
2
3
4
1
2
3
4
References
University of California – Irvine. “Mushroom Dataset”, May
1989. http://archive.ics.uci.edu/ml/datasets/Mushroom
Grayson Leonard, Matt Schartman. “Classifying Edibility of
Mushrooms”, June 2012.
Min-Ling Zhang, Zhi-Hua Zhou. “A K-Nearest Neighbor Based
Algorithm for Multi-label Classification”.
Wikipedia. “Naïve Bayes Classifier”.
http://en.wikipedia.org/wiki/Naive_Bayes_classifier
Questions?