Comparative Study of Machine Learning Algorithms on Binary Dataset
| Autor: | Rajat Puri, Digvijay Patil |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
0303 health sciences
business.industry Computer science 05 social sciences Binary number Machine learning computer.software_genre 03 medical and health sciences 0502 economics and business General Earth and Planetary Sciences Artificial intelligence business computer 050203 business & management 030304 developmental biology General Environmental Science |
| Zdroj: | International Journal of Advanced Research in Science, Communication and Technology. :137-147 |
| ISSN: | 2581-9429 |
| DOI: | 10.48175/ijarsct-887 |
| Popis: | In the world of Machine Learning, there are a lot of machine learning models to choose from for classification and decision making. Choosing the right model requires one to take in consideration various metrics like accuracy, computation time, F1 score, etc. This paper aims at comparing the performance of various such machine learning models. We use the diabetes symptoms dataset for this study. This dataset contains sixteen factors that have been seen in diabetic patients that includes age, gender, obesity, etc. The emphasis is on comparing various Machine Learning models including likes of Decision Trees, Neural Networks, etc. Decision Trees gave the best results with an accuracy of 96% and a computation time of 0.0288 seconds. Gaussian Naive Bayes was the least accurate with an accuracy of 89% and a computation time of 0.39 seconds. The great performance of Decision Trees can be attributed to the fact that the independent factors and output classes are binary and hence classification is easier and more accurate for decision trees. This paper aims at highlighting the difference in performance of various Machine Learning models based on the type of dataset used. Each model has a dataset that is most suited to it for the best possible performance. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|