Comparison of Multi-class Algorithms for Letter Recognition

In this experiment I create and compare the following five multi-class classification algorithms: decision tree, decision jungle, logistic regression, neural networks and two-class support vector machine. As shown in the results table, the neural network algorithm performed best with a macro recall of 0.930 and a macro precision of 0.933. **Data** The dataset for this experiment is the letter image recognition data from the UCI repository. The first column is the label, which identifies each row as one of 26 letters, A-Z. The remaining 16 columns are feature columns. The dataset contains 20000 instances. Description and other details about the data can be found at [http://archive.ics.uci.edu/ml/machine-learning-databases/letter-recognition/letter-recognition.names.][1] [1]: http://archive.ics.uci.edu/ml/machine-learning-databases/letter-recognition/letter-recognition.names.