ANALISIS SENTIMENT MASYARAKAT TERHADAP PENDIDIKAN 4.0 MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER

Raden Mas Haiqal Agnisfatullah, NIM.: 15650059 (2021) ANALISIS SENTIMENT MASYARAKAT TERHADAP PENDIDIKAN 4.0 MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (ANALISIS SENTIMENT MASYARAKAT TERHADAP PENDIDIKAN 4.0 MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER) 15650059_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf Download (1MB) | Preview
	Text (ANALISIS SENTIMENT MASYARAKAT TERHADAP PENDIDIKAN 4.0 MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER) 15650059_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf Restricted to Registered users only Download (3MB)

Abstract

Industrial Revolution 4.0 was an era in which technology became the main driver in human life in all fields. In this era, it has evolved in a virtual world that collects human life. The term is known as Internet Of Things (IoT). This era has affected all aspects of life, one of which is the world of Education. Resulting in the world of education follows the development of information technology and artificial intelligence technology so that the learning process is based on cyber systems. Twitter is an online social networking and microblogging service that is very popular among various circles. Education 4.0 is one of the most interesting topics of tweet unrest on Twitter and many views are pros and cons. Therefore, research on Sentiment Analysis on Education 4.0 was conducted. Analyze sentiment with Education 4.0 keywords with the Naive Bayes Classifier method. This method is an algorithm used to look up probability and statistical values. Data used as many as 1894 tweets for training data and as many as 63 tweets as test data to be classified automatically. This research was conducted using two weighting models, namely TF and IDF and using two accuracy models namely K-Fold Cross Validation and Leave One Out Cross validation. Weighting term frequency using K-Fold Cross Validation accuracy got a score of 70.31% and in Leave One Out accuracy got a score of 68.82%. While in weighting term frequency-invers document frequency with accuracy K-Fold Cross Validation got a score of 67.19% and in leave One Out accuracy got a score of 45.21%.

Share this knowledge with your friends :

Actions (login required)

View Item