PENGARUH SMOTE (SYNTHETIC MINORITY OVERSAMPLING TECHNIQUE) UNTUK MENGATASI IMBALANCE DATA PADA ANALISIS SENTIMEN MENGGUNAKAN ALGORITMA K-NEAREST NEIGHBORS

FATIYA, RAISHA and Yusliani, Novi and Marieska, Mastura Diana (2021) PENGARUH SMOTE (SYNTHETIC MINORITY OVERSAMPLING TECHNIQUE) UNTUK MENGATASI IMBALANCE DATA PADA ANALISIS SENTIMEN MENGGUNAKAN ALGORITMA K-NEAREST NEIGHBORS. Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021381823128.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_55201_09021381823128_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (10MB) | Request a copy
Preview	Text RAMA_55201_09021381823128_0008118205_0021038607_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB) | Preview
	Text RAMA_55201_09021381823128_0008118205_0021038607_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (332kB) | Request a copy
	Text RAMA_55201_09021381823128_0008118205_0021038607_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (209kB) | Request a copy
	Text RAMA_55201_09021381823128_0008118205_0021038607_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021381823128_0008118205_0021038607_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (215kB) | Request a copy
	Text RAMA_55201_09021381823128_0008118205_0021038607_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (48kB) | Request a copy
	Text RAMA_55201_09021381823128_0008118205_0021038607_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (120kB) | Request a copy
	Text RAMA_55201_09021381823128_0008118205_0021038607_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy

Abstract

The problem of imbalanced data is one of the most often problems that appears in machine learning field. A data is said to be imbalanced if the dataset is divided into a majority class and a minority class. The majority class has far more data than the minority class so that the classification results will be biased towards the majority class. Synthetic Minority Oversampling Technique (SMOTE) can be used to overcome the problem of imbalanced data that occurs. SMOTE will overcome this problem by forming synthetic data on the minority class so that the number of minority class data is balanced with the majority class. This research will carry out the process of classifying sentiment analysis using the K-Nearest Neighbors algorithm. The results of the evaluation in this study resulted in an increase in the average values of accuracy, precision, recall, and f-measure of about 8%, 4%, 10%, and 10% respectively on KNN+SMOTE. This research shows that SMOTE can be used to overcome the problem of imbalanced data and can improve the performance results of the classification model.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Sentiment Analysis, Imbalanced Data, Natural Language Processing, Synthetic Minority Oversampling Technique, K-Nearest Neighbors
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Raisha Fatiya
Date Deposited:	13 Jan 2022 03:55
Last Modified:	13 Jan 2022 03:55
URI:	http://repository.unsri.ac.id/id/eprint/60971

Actions (login required)

View Item