In building a classification model, variables containing low predictive information are sometimes used. This can increase the bias on classification. Weight of Evidence (WoE) and Information Value (IV) provide a good theoretical foundation to explore, filtering, and transforming variables in binary classification. The value of IV can help measure the predictive power possessed by a variable in separating binary classes. This research implements this framework to screen 24 predictor variables that will be used in the svm classification model to improve the evaluation of the food insecure household classification model. We use the National Socioeconomic Survey by the Indonesian Central Bureau of Statistics in 2020 for West Java Province and 2021 for East Java Province to produce a classification model. The results of this study showed that WOE was able to improve the model evaluation value from the AUC value of 0.81 to 0.83 for West Java Province and the AUC value of 0.58 to 0.66 for East Java Province.
During the current pandemic, the development of information and communication technology is very much needed in online learning. The problem in this study is how to describe the level of problem-solving abilities carried out by students in solving math problems on number material during online learning. This study aims to identify the difficulties of seventh grade students of SMP Nurul Chotib Wringinagung in learning mathematics on numbers. The type and approach in this research is descriptive research with a qualitative approach. The data collection techniques in this study were tests, interviews and questionnaires, while data processing used the percentage formula P = × 100%. From the results of the study, it was concluded that the difficulty of grade VII students of SMP Nurul Chotib Wringinagung in number material was the lack of understanding of students in solving problem solving problems, often incorrectly using formulas in solving problems, as well as difficulty distinguishing between planning and implementing plans. Keywords: difficulties problem solvig, numbers, online learning
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.