Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/169483Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Sukoco, Heru | - |
| dc.contributor.advisor | Mushthofa | - |
| dc.contributor.author | Shafa, Dhianita | - |
| dc.date.accessioned | 2025-08-15T09:02:39Z | - |
| dc.date.available | 2025-08-15T09:02:39Z | - |
| dc.date.issued | 2025 | - |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/169483 | - |
| dc.description.abstract | Internet merupakan bagian penting dalam mendukung aktivitas akademik dan non-akademik di lingkungan kampus. Namun, penggunaan internet yang berlebihan untuk aktivitas non-akademik dapat memengaruhi performa akademik. Penelitian ini bertujuan untuk mengklasifikasikan lalu lintas jaringan IPB menggunakan algoritma XGBoost serta menganalisis tren penggunaannya berdasarkan hasil klasifikasi tersebut. Data yang digunakan berupa log firewall Palo Alto yang dikumpulkan selama dua pekan pada jam sibuk. Data diproses melalui pembersihan, pengategorian ke dalam tujuh kategori trafik, rekayasa fitur, serta pembagian data untuk pelatihan, validasi, serta pengujian. Model XGBoost dibangun dengan tuning lima parameter utama, menghasilkan akurasi 94,89% dan F1-score 94,72%. Hasil klasifikasi menunjukkan bahwa trafik didominasi oleh kategori Social Network yang mencapai 39,04% dengan total volume 8 TB, diikuti oleh Infrastructure sebesar 24,14% atau sekitar 4,9 TB. Sementara itu, kategori Education & Productivity hanya menyumbang 6,33% dengan total volume 1,2 TB. Temuan ini menunjukkan adanya ketidakseimbangan penggunaan internet di lingkungan kampus, sehingga diperlukan pengelolaan bandwidth yang lebih terarah untuk mendukung prioritas akademik. Penelitian ini juga memberikan dasar bagi pengembangan sistem pemantauan jaringan berbasis machine learning. | - |
| dc.description.abstract | The internet plays a vital role in supporting both academic and non-academic activities on campus. However, excessive use of the internet for non-academic purposes can negatively impact academic performance. This study aims to classify network traffic at IPB using the XGBoost algorithm and analyze usage trends based on the classification results. The data used consists of Palo Alto firewall logs collected over two weeks during peak hours. The data underwent cleaning, categorization into seven traffic categories, feature engineering, partitioning for training, validation, and testing. The XGBoost model was built by tuning five main parameters, resulting in an accuracy of 94.89% and an F1-score of 94.72%. The classification results show that traffic is dominated by the Social Network category, which accounts for 39.04% with a total volume of 8 TB, followed by Infrastructure at 24.14% or approximately 4.9 TB. Meanwhile, the Education & Productivity category only contributed 6.33% with a total volume of 1.2 TB. These findings indicate an imbalance in internet usage on campus, highlighting the need for more targeted bandwidth management to support academic priorities. This study also provides a foundation for the development of machine learning-based network monitoring systems. | - |
| dc.description.sponsorship | null | - |
| dc.language.iso | id | - |
| dc.publisher | IPB University | id |
| dc.title | Klasifikasi Lalu Lintas Jaringan Menggunakan Machine Learning untuk Analisis Tren Penggunaan Internet di IPB | id |
| dc.title.alternative | Network Traffic Classification Using Machine Learning for Internet Usage Trend Analysis at IPB University | - |
| dc.type | Skripsi | - |
| dc.subject.keyword | penggunaan internet | id |
| dc.subject.keyword | XGBoost | id |
| dc.subject.keyword | XGBoost | id |
| dc.subject.keyword | klasifikasi lalu lintas | id |
| dc.subject.keyword | pengelolaan bandwidth | id |
| dc.subject.keyword | bandwidth management | id |
| dc.subject.keyword | internet usage | id |
| dc.subject.keyword | traffic classification | id |
| Appears in Collections: | UT - Computer Science | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| cover_G6401211016_53beb5f335db4c5d92eb33893d3144e2.pdf | Cover | 2.32 MB | Adobe PDF | View/Open |
| fulltext_G6401211016_0472c6f7d4994a3e9eb5ceb5d245ebd1.pdf Restricted Access | Fulltext | 4.17 MB | Adobe PDF | View/Open |
| lampiran_G6401211016_fc7f4e9dd7d44b48aee29379a5dc0c1d.pdf Restricted Access | Lampiran | 2.51 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.