Claim Missing Document
Check
Articles

Found 12 Documents
Search

SISTEM INFORMASI MANAJEMEN SURAT PADA FITUR SURAT DINAS DI BADAN PUSAT STATISTIK KABUPATEN PRINGSEWU Ridho Sholehurrohman; Igit Sabda Ilman; Mita Anggraaeni; rahman taufik; Muhaqiqin
Jurnal Komputasi Vol. 13 No. 1 (2025)
Publisher : Jurusan Ilmu Komputer Fakultas MIPA Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/komputasi.v13i1.280

Abstract

The development of information technology is getting faster so that there are many uses of computer-based programs that help the performance of an organization, both offices and companies. One of the organizations that utilizes information technology is the BPS Pringsewu. In data processing, BPS really determines how far the level of accuracy and precision of the statistical data produced. Also in its development, the BPS Pringsewu made use of the use of computer technology by using a mail management information system. In the mail management information system it is effectively and efficiently to assist the process of inputting data encountered and possibly when errors (fails) occur when typing is done repeatedly. From the results of the research conducted, it can be concluded that the Mail Management Information System on Official Letter Features is a website-based information system that has two levels of users, namely admin and employees. This system is expected to facilitate agencies in the process of storing incoming and outgoing mail data, and facilitate the process of automatically assigning outgoing mail numbers. Furthermore, from the results of black-box testing and User Acceptance Testing was found that this system has user friendly features making it easier for users to use the system, seen in the average test questionnaire getting 95.68%.
Klasifikasi Dokumen Akademik Berbasis XGBoost untuk Pemetaan Tujuan Pembangunan Berkelanjutan (SDGs) di Universitas Lampung Rahman Taufik; Arsitoteles; Candra Wijaya; Rakhmat Herlambang
Jurnal Komputasi Vol. 13 No. 2 (2025)
Publisher : Jurusan Ilmu Komputer Fakultas MIPA Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/komputasi.v13i2.329

Abstract

Pemetaan kontribusi institusi pendidikan tinggi terhadap Sustainable Development Goals merupakan tantangan krusial untuk akuntabilitas global dan capaian World Class University. Meskipun model-model canggih rentan terhadap overfitting dan menuntut sumber daya komputasi besar pada data yang tidak seimbang, penelitian ini mengeksplorasi algoritma XGBoost sebagai solusi efisien untuk klasifikasi SDGs pada dokumen akademik universitas. Penelitian ini menggunakan dataset sebanyak 148136 dokumen, diproses dengan TF−IDF, dan dioptimasi dengan hyperparameter tuning serta class sample weighting untuk mitigasi imbalance. Hasil evaluasi menunjukkan model yang stabil dengan accuracy 0.92, precision 0.92, recall 0.89, dan F1−score 0.90 pada dataset uji. Meskipun kinerja agregat tinggi, analisis log loss dan confusion matrix mengindikasikan adanya overfitting lokal pada kategori minoritas, yang menyebabkan recall rendah di kelas-kelas tersebut. Secara keseluruhan, model XGBoost terbukti valid sebagai alat ukur efektif untuk memetakan kontribusi universitas terhadap SDGs, sekaligus memberikan panduan strategis berbasis data untuk mengidentifikasi celah dan mendorong keseimbangan capaian WCU