Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : Journal of Information System and Application Development

Analisis sentimen kebijakan masuk sekolah pukul lima pagi menggunakan algoritma Naïve Bayes Hoar, Wilhelmina Sonya; Zubair, Anis; Muflikhah, Lailil
Journal of Information System and Application Development Vol. 2 No. 1 (2024): March 2024
Publisher : Universitas Merdeka Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26905/jisad.v2i1.10935

Abstract

Education is very important to build a quality young generation that can advance the nation. To improve the quality of education, the government of East Nusa Tenggara implemented a five am school entry policy. However, this policy has caused pros and cons in the community. People are becoming more active expressing their opinions through social media. Criticism of this policy is reflected in comments appearing on social media, especially on Twitter. Therefore, it's important to conduct sentiment analysis to know how many positive and negative responses to this policy. In this research, sentiment analysis was carried out on search results for tweets with the keyword “sekolah jam 5 di NTT” in the time period from February to March 2023. A total of 777 tweets were obtained with 24 positive sentiments and 753 tweets with negative sentiments. The data was then processed and analyzed using the Naïve Bayes algorithmThis research obtained accuracy results of 97% with a negative sentiment precision value of 98% and a positive sentiment precision value of 50%. In addition, the recall and f1-score values for negative sentiment are greater than positive sentiment, indicating that more people do not agree with the policy.
Perbandingan prediksi jumlah penjualan kapas menggunakan metode dekomposisi aditif dan multiplikatif Kristuadji, Yesaya Arya Danar; Zubair, Anis
Journal of Information System and Application Development Vol. 2 No. 2 (2024): September 2024
Publisher : Universitas Merdeka Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26905/jisad.v2i2.11068

Abstract

PT Taruna Kusuma Purinusa is a company that focuses on the beauty cotton industry with the Selection Cotton brand label. Their product sales are spread in various cities including Malang City. This study aims to determine fluctuations and predict the amount of sales of cotton PT Taruna Kusuma Purinusa Malang and simultaneously compare the two methods of Decomposition, namely Multiplicative Decomposition and Additive Decomposition. The data is secondary on cotton sales from January 2019 to May 2023. The research procedure begins by analyzing the components of the decomposition, namely the trend (𝑇), seasonal (𝑆), cyclical (𝐶) and random (𝐼) components, then multiplying the value of these components. The prediction results show that cotton sales from June to December 2023 are as follows: 1,759,864 in June, 1,691,855 in July, 1,744,614 in August, 1,720,060 in September, 1,746,516 in October, 1,850,913 in November, and 1,893,669 in December. In addition, the proper forecasting method used in the Cotton Sales data at PT Taruna Kusuma Purnisua Malang is the additive decomposition method.
Analisis prediktif perubahan nilai profit berdasarkan klasifikasi pengguna pada usaha jasa logistik Putri, Sephia Dwi Arma; Zubair, Anis
Journal of Information System and Application Development Vol. 1 No. 1 (2023): March 2023
Publisher : Universitas Merdeka Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26905/jisad.v1i1.9865

Abstract

The delivery of logistics services continues to experience a rapid increase in line with the high sales of goods through e-commerce. According to thepredetermined standards, each cargo expedition outlet is required to meet the monthly shipment achievement target (tonnage) to avoid fines and an increase in the target rate. This study aims to find out how to anticipate failure to achieve targets, so as to prevent additional operating expenses and profit instability. In this study, quantitative analysis was carried out using two algorithms. The Naïve Bayes algorithm is used to classify service user categories. The results showed that the industrial category contributed greatly to the delivery of cargo expeditions with a total tonnage percentage of 69.33%, while the remaining 30.67% was included in the individual category. Furthermore, the Multiple Linear Regression algorithm is used to predict profit values based on category classes. Predictions were made from October to December which resulted in an increase in profit for the individual category and a decrease in profit for the industrial category. Recommendations that can be made include royalty rewards for customers, brand awareness to attract new customers, as well as business cooperation with MSMEs and other business actors around them.