Zobraziť v EDS

Penggunaan Algoritma C4.5 untuk Prediksi Waktu Penyelesaian Gangguan pada Layanan Internet.

Uložené v:
Podrobná bibliografia
Názov: Penggunaan Algoritma C4.5 untuk Prediksi Waktu Penyelesaian Gangguan pada Layanan Internet. (Indonesian)
Autori: Fransdela, Rion
Zdroj: Journal of Syntax Literate; Sep2025, Vol. 10 Issue 9, p7083-7094, 12p
Predmety: PREDICTION models, TURNAROUND time, NETWORK failures (Telecommunication), DECISION trees, CUSTOMER service management, DATA mining, CUSTOMER satisfaction
Geografický termín: INDONESIA
Abstract (English): Internet service disruptions can significantly affect customer satisfaction and a company's reputation. PT. Xita Telekomunikasi Indonesia currently relies on manual methods to estimate disruption resolution time, often leading to uncertainty in customer communication. This research aims to develop a predictive model for estimating internet service disruption resolution time using the C4.5 algorithm. Historical disruption data was collected and processed through cleaning and attribute transformation stages, followed by classification modeling using Python. The results show that the model effectively classifies resolution times into three categories: Fast (=2 hours), Medium (2-5 hours), and Long (>5 hours), achieving an accuracy of 70%. The "Medium" category showed the best performance with an F1-score of 0.84, followed by "Long" at 0.73, while "Fast" had a relatively low F1-score of 0.22. Practically, this model can be integrated into PT. Xita Telekomunikasi Indonesia to provide more accurate and consistent estimates of turnaround times, thereby increasing transparency and customer trust. Theoretically, this research enriches the application of data mining in the field of information technology service management, especially in the context of providing internet services in Indonesia. This prediction model can assist the company in providing more accurate time estimates to customers, thereby improving operational efficiency and service quality. [ABSTRACT FROM AUTHOR]
Abstract (Indonesian): Gangguan layanan internet dapat berdampak signifikan terhadap kepuasan pelanggan dan reputasi perusahaan. PT. Xita Telekomunikasi Indonesia masih menggunakan pendekatan manual untuk memperkirakan durasi penyelesaian gangguan, yang seringkali menimbulkan ketidakpastian informasi. Penelitian ini bertujuan membangun model prediksi estimasi waktu penyelesaian gangguan layanan internet menggunakan algoritma C4.5. Data gangguan historis dikumpulkan dan diproses melalui tahapan pembersihan serta transformasi atribut, kemudian diolah menggunakan Python untuk membentuk model klasifikasi. Hasil evaluasi menunjukkan bahwa model mampu mengelompokkan estimasi waktu ke dalam tiga kategori: Cepat (=2 jam), Sedang (2-5 jam), dan Lama (>5 jam), dengan akurasi mencapai 70%. Kategori "Sedang" menunjukkan performa terbaik dengan f1-score sebesar 0.84, disusul kategori "Lama" (0.73), sedangkan kategori "Cepat" memiliki f1-score rendah yaitu 0.22. Secara praktis, model ini dapat diintegrasikan ke dalam sistem layanan pelanggan PT. Xita Telekomunikasi Indonesia untuk memberikan estimasi waktu penyelesaian yang lebih akurat dan konsisten, sehingga meningkatkan transparansi dan kepercayaan pelanggan. Secara teoretis, penelitian ini memperkaya penerapan data mining dalam bidang manajemen layanan teknologi informasi, khususnya dalam konteks penyediaan layanan internet di Indonesia. Model ini dapat membantu perusahaan memberikan estimasi waktu yang lebih akurat kepada pelanggan, sehingga meningkatkan efisiensi operasional dan kualitas layanan. [ABSTRACT FROM AUTHOR]
Copyright of Journal of Syntax Literate is the property of Ridwan Institute and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Databáza: Complementary Index
Popis
Abstrakt:Internet service disruptions can significantly affect customer satisfaction and a company's reputation. PT. Xita Telekomunikasi Indonesia currently relies on manual methods to estimate disruption resolution time, often leading to uncertainty in customer communication. This research aims to develop a predictive model for estimating internet service disruption resolution time using the C4.5 algorithm. Historical disruption data was collected and processed through cleaning and attribute transformation stages, followed by classification modeling using Python. The results show that the model effectively classifies resolution times into three categories: Fast (=2 hours), Medium (2-5 hours), and Long (>5 hours), achieving an accuracy of 70%. The "Medium" category showed the best performance with an F1-score of 0.84, followed by "Long" at 0.73, while "Fast" had a relatively low F1-score of 0.22. Practically, this model can be integrated into PT. Xita Telekomunikasi Indonesia to provide more accurate and consistent estimates of turnaround times, thereby increasing transparency and customer trust. Theoretically, this research enriches the application of data mining in the field of information technology service management, especially in the context of providing internet services in Indonesia. This prediction model can assist the company in providing more accurate time estimates to customers, thereby improving operational efficiency and service quality. [ABSTRACT FROM AUTHOR]
ISSN:25410849