NobleBlocks

Research indexJournal article

ການຄາດຄະເນບນຫາສາຍໄຍແກວລະບບໃຕດນຂອງສນສສານກອງທບ, ກະຊວງປອງກນປະເທດໂດຍນາໃຊເຕກນກການຮຽນຮຂອງຄອມພວເຕ

Souvanath, Inthasone, Sonevilai

Lao Journal of Science and Technology31 Jul 2025Open access

NobleIDark:/48914/w1/20261105/C627E76EDOI10.71026/ljst.2024.01007

Abstract

Currently, information technology and telecommunication systems are so importance in human living and play an important role in all activities such as business, education and entertainment. For national defense task Telecommunication and information systems by underground fiber network is still hard to plan and unpredictable of effectiveness maintenance and repair. Therefore, this study aims to use Machine learning techniques to analyze and predict the quality of the fiber optic cables due to further analysis and planning. This research is using the theory of Machine learning, The datasets obtained from OTDR collected from 2021-2023. The Datasets are 2576 samples and using Three Models: K-Nearest Neighbors (KNN), Support Vector Machine (SVM) and Logistic Regression (LR) then comparing the most accuracy of the models and selecting the best model to develop as a platform. After using those Tree models of experimental results show that K-Nearest Neighbors (KNN) model has the highest accuracy of 98% compared to the SVM model of 94% and LR 97% with the value of (k) : 3. And KNN model to be more accurate, I adjusted the (k) value to 3 Number: 3, 7 and 9. That shows that the correct value is still the K value equal to: 3. So the KNN model with a value of (k) : 3 is correct and that model to use.

Discussion

Citations

More

Suggest a correction or report