Search

Showing total 3 results
Start Over Topic algorithms Topic cart algorithms Publication Year Range Last 50 years
3 results

1. Design of laboratory room monitoring system using multi-sensor and CART algorithm.

Author

Zakiy, Deza Achmad and Nugraha, I. Gde Dharma

Subjects
*CART algorithms, *ALGORITHMS, *COLLEGE environment, *REGRESSION trees, *ENVIRONMENTAL monitoring, *TESTING laboratories, *MICROCONTROLLERS, *LABORATORIES
Abstract

ISO 14001 is a set of standards to help organizations for constructing an environmentally building or rooms. In the campus environment, the ISO 14001 is used as the guide for creating an environmentally laboratory. Based on this ISO, the environmental condition of the laboratory need to be checked periodically. Recently, IoT devices were developed as the tools to monitor the ecological status of the laboratory. However, the recent study of IoT-based monitoring is used only for collecting the data. The data is being processed manually on the server. In our research, we try to implement machine learning for the IoT-based monitoring tools to improve the performance and gives the capability to respond to any condition of the laboratory. In this paper, we discussed our proposed design of IoT-based monitoring devices. We used a microcontroller module called NodeMCU ESP8266 to build an efficient monitoring system. ESP8266 is a type of microcontroller board equipped with a WiFi module to make it possible to design a system that can send data from multiple sensors (temperature, humidity, light intensity, CO2 concentration) to be displayed and sent to the database server using the WiFi module. The data collected in the database will be processed using machine learning by the Classification and Regression Tree (CART) algorithm and then implemented to the microcontroller as embedded machine learning to detect impending early threats and provide early warnings. With this method, it has been found that the CART algorithm provides a speedy processing time with training and testing time of 0.5 seconds and 0.06 seconds, the precision of 0.999154, recall of 0.999946, and f1-score of 1.0. We also design our device to be compact to be placed anywhere inside the laboratory and consume low-power. [ABSTRACT FROM AUTHOR]

Published
2021
Full Text
View/download PDF

2. C_CART: An instance confidence-based decision tree algorithm for classification.

Author

Yu, Shuang, Li, Xiongfei, Wang, Hancheng, Zhang, Xiaoli, and Chen, Shiping

Subjects
*CART algorithms, *DECISION trees, *CLASSIFICATION algorithms, *REGRESSION trees, *ALGORITHMS, *CONFIDENCE
Abstract

In classification, a decision tree is a common model due to its simple structure and easy understanding. Most of decision tree algorithms assume all instances in a dataset have the same degree of confidence, so they use the same generation and pruning strategies for all training instances. In fact, the instances with greater degree of confidence are more useful than the ones with lower degree of confidence in the same dataset. Therefore, the instances should be treated discriminately according to their corresponding confidence degrees when training classifiers. In this paper, we investigate the impact and significance of degree of confidence of instances on the classification performance of decision tree algorithms, taking the classification and regression tree (CART) algorithm as an example. First, the degree of confidence of instances is quantified from a statistical perspective. Then, a developed CART algorithm named C_CART is proposed by introducing the confidence of instances into the generation and pruning processes of CART algorithm. Finally, we conduct experiments to evaluate the performance of C_CART algorithm. The experimental results show that our C_CART algorithm can significantly improve the generalization performance as well as avoiding the over-fitting problem to a certain extend. [ABSTRACT FROM AUTHOR]

Published
2021
Full Text
View/download PDF

3. مدل بندی و داده کاوی داده های جهانی بیماران ویروس کووید ۱.

Author

مصطفی بسکابادی and مهدی دوست پرست

Subjects
*CART algorithms, *COVID-19, *RHINORRHEA, *RECEIVER operating characteristic curves, *DECISION trees, *COUGH, *SNEEZING
Abstract

Introduction: Data mining techniques, including decision tree algorithms can be used for modeling and identifying those at risk of developing COVID-19. The main goal of this study is estimating the risk of death of patients due to COVID-19, using the classification and regression tree (CART) algorithm based on the observed effective factors. Methods: This paper is an analytical study and the data of all patients with COVID-19 registered on the Kaggle site through Johns Hopkins University, was extracted. There was a total of 26,031 records from various countries. Data analysis was performed using JMP statistical software version 13. In the modeling section, decision tree algorithm and CART model were used. Results: The results of the classification and regression tree showed that among quantitative variables, age, the interval between hospitalization and result, the interval between onset of symptoms and test result, and the interval between hospitalization and test result, and the qualitative variable of gender were the most important factors affecting the outcome of patient treatment, respectively. According to the analysis of words, fever, cough, sore throat, fatigue, weakness, headache, chills, and runny nose, respectively, were the most common symptoms among patients with this disease. Conclusion: The accuracy of the fitted model was shown to be 94.1% for experimental data and 91.1% for educational data using the area under the receiver operating characteristic (ROC) curve. [ABSTRACT FROM AUTHOR]

Published
2020