A Survey on News Classification Techniques

Katari, Rohan; Bala, Myneni Madhu

doi:10.1109/iccsea49143.2020.9132866

Cited by 6 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Klasifikasi adalah pendekatan supervised learning yang beberapa algoritmanya seperti Naïve Bayes, Decision Tree, Neural Network, SVM, Random Forest [1] [2]. Kasus klasifikasi dapat ditemukan diberbagai domain [3]- [5]. Salah satu pemanfaatan dalam klasifikasi adalah penentuan jenis cuaca.…”

Section: Pendahuluanunclassified

Klasifikasi Cuaca Provinsi Dki Jakarta Menggunakan Algoritma Random Forest Dengan Teknik Oversampling

Hamami

Dahlan

2022

JTI

View full text Add to dashboard Cite

Nowadays Indonesia experiences extreme weather changes that cause many disasters such as floods, fires, landslides and storms. The type of weather depends on many factors such as temperature, humidity, wind direction and others. Some human activities depend on changes in weather such as in the agricultural sector, plantations, aviation, highlands and beaches. Weather prediction is important to understand extreme weather changes based on weather factors. This research adopts ensemble learning which is able to perform weather classification. The algorithm used is Random Forest combined with oversampling technique to handle the uneven amount of data from each weather class. Some of the weather categories classified are Sunny, Sunny Cloudy, Cloudy, Heavy Cloudy, Local Rain, Light Rain, Moderate Rain and Thunderstorms. The experimental results show that the Random Forest model achieves accuracy of 70%. The oversampling technique used is the Synthetic Minority Over-sampling Technique (SMOTE) method. With the combination of SMOTE the predictions of each minority class can be increased by average of 50%..

show abstract