E-commerce merupakan aplikasi yang digunakan untuk melakukan segala aktivitas jual-beli secara online beberapa masalah transaksi jual-beli online terutama pada shopee terkadang banyak kendala. Ulasan dari pengguna sering digunakan sebagai alat yang efektif dan efisien untuk mencari informasi tentang suatu produk atau layanan. Pengguna internet sering melihat ulasan suatu produk sebelum menggunakannya karena ulasan dari pengguna lain dapat memberikan informasi terkini. Pada penelitian ini penulis menggunakan dua (2) jenis metode algoritma yakni K-Nearst Neighbor dan Naive Bayes Classifier. Penulis menggunakan dua (2) metode untuk mengetahui tingkat akurasi terbaik yang di hasilkan. Penulis menggunakan algoritma K-Nearst Neighbor karena pada penelitian sebelumnya terdapat hasil yang bagus. Juga menggunakan algoritma Naïve Bayes Classifier karena metode ini merupakan metode yang popular dan lebih sering digunakan.algoritma K-Nearst Neighbor pada aplikasi shopee menunjukkan bahwa tingkat akurasi 55.00% dimana jumlah data untuk sentimen negatif sebanyak 100 ulasan, dan sentimen positif sebanyak 100 ulasan dari data testing yang digunakan sebanyak 200 data atau 20% dalam metode K-Nearst Neighbor. Lalu tingkat akurasi mencapai 80,00% dimana jumlah data untuk sentimen negatif sebanyak 99 ulasan dan sentimen positif sebanyak 101 ulasan dari data testing yang digunakan sebanyak 200 atau 20% dengan metode Naïve Bayes Classifier. Dalam penelitian ini, dapat diketahui bahwa metode Naïve Bayes Classifier adalah metode yang tingkat akurasinya lebih tinggi dibandingkan K-Nearst Neighbor dengan tingkat akurasi sebesar 80,00%.

With the development of today's modern era, publik service is an important and very necessary thing because it is one of the benchmarks for seeing publik trust and satisfaction with the services provided by an agency. One of the agencies that carries out publi services is the Balai Pengujian Standar Instrumen Tanaman Industri dan Penyegar (BPSI TRI), a government agency under the Ministry of Agriculture. There are a lot of people who will receive services in 2023. Therefore, publik service officers find it difficult to determine publik satisfaction in order to optimize the services provided. To determine community satisfaction, data mining calculations were carried out using the K-Means clustering algorithm method with Community Satisfaction Index (IKM) data in 2023 using 3 (three) categories including unsatisfactory (C1), satisfactory (C2) and very satisfactory) and 2 attributes, namely the behavior of service officers (U7) as well as handling complaints, suggestions and input (U8) then carried out calculations using Microsoft Excel and got the results that C1 (unsatisfactory) 14 respondents, C2 (satisfactory) 39 respondents and C3 (very satisfactory) 98 respondents. Meanwhile, from the results of calculations using python testing, the results showed that C1 (unsatisfactory) was 9 respondents, C2 (satisfactory) was 39 respondents and C3 (very satisfactory) was 103 respondents.

The main purpose of this research paper is to assess the prevalence, diagnosis, and management of common cattle diseases in different countries, including Jordan, Pakistan, Uganda, Korea, Bangladesh, and Europe. Our dataset includes 163 cases and 123 detailed symptoms; this research identifies patterns of symptoms with great accuracy. The accuracy of the dataset is 98%. The main diseases in cattle include digestive disorders, osteodystrophy, tick-borne diseases, and lumpy skin disease. Two types of tools were used: innovative diagnostic tools, like fuzzy logic models, and a diagnostic decision support tool (DDST). This tool performs disease detection and management. The findings demonstrate the importance of accurate diagnosis in vaccination programs and biosecurity measures in order to adequately measure economic losses and improve livestock health.

Employee attendance is a crucial factor in human resource management as it affects productivity and operational efficiency. However, the recording and analysis of employee attendance often encounter challenges, particularly in terms of the accuracy and effectiveness of the systems used. This study aims to develop an employee attendance classification model using the Gradient Boosted Trees algorithm to improve the accuracy of grouping attendance categories such as Present, Permission, Sick, Leave, and Absent into attendance level categories: High, Medium, and Low. The research method includes collecting employee attendance data throughout the year 2024. The model evaluation is carried out using metrics such as accuracy, precision, recall, and the confusion matrix. The results indicate that the developed model achieves an accuracy of 100.00%, with a mean precision of 100.00% and a mean recall of 100.00%.