Template Credit: Adapted from a template made available by Dr. Jason Brownlee of Machine Learning Mastery.
SUMMARY: This project aims to construct a predictive model using various machine learning algorithms and document the end-to-end steps using a template. The Sign Language MNIST dataset is a multi-class classification situation where we attempt to predict one of several (more than two) possible outcomes.
INTRODUCTION: The original MNIST image dataset of handwritten digits is a popular benchmark for image-based machine learning methods. The Sign Language MNIST is presented here and follows the same CSV format with labels and pixel values in single rows to stimulate the community to develop more drop-in replacements. The American Sign Language letter database of hand gestures represent a multi-class problem with 24 classes of letters (excluding J and Z, which require motion).
The dataset format is patterned to match closely with the classic MNIST. Each training and test case represents a label (0-25) as a one-to-one map for each alphabetic letter A-Z (and no cases for 9=J or 25=Z because of gesture motions). The training data (27,455 cases) and test data (7172 instances) are approximately half the size of the standard MNIST but otherwise similar with a header row of the labels, pixel1,pixel2….pixel784 which represent a single 28×28 pixel image with grayscale values between 0-255. The original hand gesture image data represented multiple users repeating the gesture against different backgrounds.
ANALYSIS: The average performance of the machine learning algorithms achieved an accuracy benchmark of 96.38%. Two algorithms (Extra Trees and Random Forest) produced the top accuracy metrics after the first round of modeling. After a series of tuning trials, the Extra Trees model turned in an accuracy metric of 99.61%. When configured with the optimized parameters, the Extra Trees model processed the validation dataset with an accuracy score of 99.83%. When we applied the Extra Trees model to the previously unseen test dataset, we obtained an accuracy score of 83.49%, which pointed to a high variance error.
CONCLUSION: In this iteration, the Extra Trees model did not appear to be suitable for modeling this dataset. We should consider experimenting another algorithm with this dataset.
Dataset Used: Sign Language MNIST Data Set
Dataset ML Model: Multi-Class classification with numerical attributes
Dataset Reference: https://www.kaggle.com/datamunge/sign-language-mnist
One source of potential performance benchmarks: https://www.kaggle.com/datamunge/sign-language-mnist
The HTML formatted report can be found here on GitHub.