<![CDATA[Predicting category and numerical data, such as working schedule data, is difficult since it necessitates a specific process. A decision tree is one of many categorization methods that can handle both category and numerical input. Scikit learn, a python library that may be used for decision trees, is one example. Although Scikit-optimized learn's CART algorithm could only handle numerical data, it did provide certain features to deal with categorical data. To forecast working schedules, this study used scikit-learn to create a decision tree model. There are 54 variables, three of which are category and one of which is numerical. A 6-depth decision tree model was created as a result of the implementation. The evaluation yielded a positive outcome, with accuracy and precision above 0.7 and 0.9, respectively. The optimal division of data is 30% validation and 70% training. In comparison to KNN, the decision tree model has higher accuracy, with decision tree accuracy exceeding 0.8 while KNN accuracy is below.]]>
Copyrights © 2022
