Please use this identifier to cite or link to this item: http://repository.potensi-utama.ac.id/jspui/jspui/handle/123456789/5686
Title: IDENTIFICATION OF PUBLIC SENTIMENT TOWARDS COVID-19 ISSUES WITH NAÏVE BAYES ALGORITHM AND LATENT SEMANTIC INDEXING (LSI)
Authors: Rosnelly, Rika
Keywords: Latent Semantic Indexing, Naive bayes, Identification.
Issue Date: Feb-2023
Publisher: KIMARA
Abstract: The Latent Semantic Indexing method is a method that is implemented in the IR system in finding and finding information based on the overall meaning (conceptual topic or meaning) of a document, not just word for word meaning. Naive Bayes is a classification using probability and statistical methods, the Naive Bayes algorithm can be used in the scientific field, one of which is predicting future opportunities based on previous experience. The results of this study are expected to provide advice actively through social media with policy makers in Indonesia regarding developing issues and impacts on society. From the combination of Latent Semantic Indexing (LSI) and Naive Bayes algorithms, the identification performance is quite good. Has a fairly high accuracy of 88.6% and an error percentage of 11.4%. Of the 150- testing data, there are 17 identification errors where the input data as test data has similarities to one data with other data.
Description: Emotions can be grouped into positive emotions and negative emotions. Human emotions can be categorized into five basic emotions, namely love, joy, sadness, anger and fear. The emotions of love and pleasure are included in positive emotions. The emotions of sadness, anger, and fear are negative emotions. In analyzing public sentiment, it is necessary to classify an opinion, both positive and negative, on Twitter. However, if you classify it manually, it will take a lot of time and effort in its implementation. Therefore, we need a way to classify an opinion more quickly and accurately. One of them is the use of Text Mining which serves to analyze or group documents or text from a large number of documents or texts. Sentiment analysis or opinion mining is the process of understanding, extracting and processing textual data automatically to obtain sentiment information contained in an opinion sentence. Sentiment analysis will classify the polarity of the text in a sentence or document to find out whether the opinion expressed in the sentence or document is positive or negative. The magnitude of the influence and benefits of sentiment analysis causes research and applications based on sentiment analysis to develop rapidly. The use of sentiment analysis can be applied to public opinion on the COVID-19 pandemic. Abidin conducted research related to the Development of an Indonesian Dictionary of Non-Standard Words Using the Latent Semantic Indexing Algorithm and Damerau Levenshtein Distance. The result of the research is to produce an effective text preprocessing module in building an Indonesian dictionary of non-standard words.
URI: http://repository.potensi-utama.ac.id/jspui/jspui/handle/123456789/5686
ISSN: 2976-3606
Appears in Collections:A Paper

Files in This Item:
File Description SizeFormat 
44. IDENTIFICATION OF PUBLIC SENTIMENT TOWARDS COVID-19.pdf3.27 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.