In Indonesia, Twitter is one of the most widely used social media platforms. Because of the diverse and frequently shifting message patterns on this social media, it is extremely challenging and time-consuming to manually identify topics from a collection of messages. Topic modeling is one method for obtaining information from social media. The model and visualization of the results of modeling topics that are discussed on social media by the Makassar community are the goals of this study. The Latent Dirichlet Allocation (LDA) algorithm is used to model and display the results of this study. The modeling results indicate that the eighth topic is the most frequently used word in a conversation. In the meantime, the 7th and 6th topics emerged as the conversation's core based on the spread of the words with the highest term frequency. The study's findings led the researchers to the conclusion that in the Makassar community's social media discussions, capitalization and visualization using the LDA method produced the words with the highest trend and the topic with the highest term frequency.
Based on data obtained from SPAN-PTKIN registrants in 2018 and 2019, the number of interested people through the invitation path who chose the study program at UIN Alauddin as the first choice was 30523 records. Analysis using the ID3 algorithm found that those who interested in the study of religions were more dominant from vocational students. While analysis using the K-Means shows the regions / regencies from which those interested in study programs of religions are spread in 35 regencies / cities. It can be concluded that the socialization of study programs of religions through the invitation path is recommended to be more focused on SMAs that are located in 33 districts / cities as identified in cluster 3. The study programs of religions are prioritized, because these study programs experienced the lowest number of registrants. It is expected that by implementing this recommended strategy, the number of interested prospective new students will draw a significant increase in the future.
Kelulusan mahasiswa merupakan salah satu tolak ukur dalam menjadikan pendidikan lanjutan sebagai strategi dewan. Saat ini sudah biasa dilacak bahwa jumlah mahasiswa baru di suatu perguruan tinggi tidak sebanding dengan jumlah lulusan setiap tahunnya sehingga peneltian ini mencoba membuat sistem yang dapat memprediksi kelulusan mahasiswa. Tujuan dari penelitian ini Untuk mengimplementasikan algoritma C.45 dalam memprediksi Kelulusan Mahasiswa dan Untuk mengetahui factor-faktor yang mempengaruhi dalam kelulusan mahasiswa tepat waktu dan tidak tepat waktu. Algoritma yang digunakan dalam penelitian ini adalah algoritma c.45 dimana Algoritma C4.5 digunakan untuk membentuk pohon keputusan yang dapat digunakan untuk membentuk pohon keputusan. Hasil dari penelitian ini adalah Mahasiswa dapat di prediksi dan dilakukan pengujian dengan menerapkan teknik data mining untuk melakukan prediksi berdasarkan data training dengan attribute SKS dan Jumlah IPK merupakan faktor yang mempengaruhi ketepatan waktu lulus dan Hasil evaluasi yang dilakukan dengan melakukan pengujian Confusion Matrix dengan data uji sebanyak 404 data mahasiswa dengan pembagian data kelulusan mahasiswa yang di prediksi lulus dengan tepat waktu sebanyak 368 sedangkan mahasiswa yang tidak lulus dengan tepat waktu terdapat 36 data mahasiswa sehingga dengan melakukan pengujian confusion matrix memperoleh hasil akurasi sebesar 85%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.