Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/170700| Title: | Perancangan dan Implementasi Modul Backend untuk Web Penapisan Senyawa Herbal Antimalaria |
| Other Titles: | Design and Implementation of a Backend Module for an Antimalarial Herbal Compound Screening Web |
| Authors: | Kusuma, Wisnu Ananta Nurhadryani, Yani NOFIANSYAH, DAFFA |
| Issue Date: | 2025 |
| Publisher: | IPB University |
| Abstract: | Malaria masih menjadi masalah kesehatan serius di Indonesia sehingga diperlukan inovasi berbasis teknologi untuk mempercepat penemuan obat alternatif. Penelitian ini berfokus pada pengembangan modul backend untuk sistem penapisan senyawa herbal antimalaria dengan arsitektur REST API sebagai penghubung antara frontend, model machine learning, dan basis data. Hasil penelitian menunjukkan bahwa modul backend berhasil menyediakan 21 endpoint API yang mencakup autentikasi, manajemen pengguna, pengelolaan data senyawa, prediksi nilai IC50, serta manajemen model. Black box testing dengan 36 skenario menunjukkan seluruh endpoint berfungsi sesuai rancangan. Stability testing terhadap 30 kali permintaan prediksi mencatat waktu respons rata-rata 380 ms dengan respons yang konsisten, tanpa adanya kegagalan sistem, sehingga layanan dinilai stabil dan andal. Volume testing menunjukkan performa optimal pada batch kecil hingga menengah (= 100 data), dengan respons <3 detik, sementara pada batch besar (> 250 data) waktu respons meningkat signifikan. Malaria remains a serious public health problem in Indonesia, thus requiring technological innovations to accelerate the discovery of alternative drugs. This research focuses on the development of a backend module for an antimalarial herbal compound screening system using a REST API architecture as a bridge between the frontend, machine learning model, and database. The results show that the backend module successfully provides 21 API endpoints covering authentication, user management, compound data management, IC50 prediction, and model management. Black box testing with 36 scenarios confirmed that all endpoints functioned as designed. Stability testing on 30 consecutive prediction requests recorded an average response time of 380 ms with consistent responses and no system failures, indicating that the service is stable and reliable. Volume testing demonstrated optimal performance for small to medium batches (= 100 data) with responses under 3 seconds, while larger batches (> 250 data) caused a significant increase in response time. |
| URI: | http://repository.ipb.ac.id/handle/123456789/170700 |
| Appears in Collections: | UT - Computer Science |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| cover_G6401211098_da72b8f1343e490e99ce2086ff6e6d9e.pdf | Cover | 433.77 kB | Adobe PDF | View/Open |
| fulltext_G6401211098_be09baaad5404cb2a6b249917d2a25c1.pdf Restricted Access | Fulltext | 1.65 MB | Adobe PDF | View/Open |
| lampiran_G6401211098_a6381824ed0044038acd26baf70263d3.pdf Restricted Access | Lampiran | 311.95 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.