Skripsi

PENERAPAN ALGORITMA K-NEAREST NEIGHBOR PADA KLASIFIKASI PETIR NBE

---

Penilaian

---

This research aims to classify lightning recording data obtained in the Palembang region in January 2020 using the K-Nearest Neighbor (KNN) algorithm implemented in the Python programming language. The raw data, initially in psdata format, was converted to Comma Separated Values (CSV) format to facilitate analysis. From a total of 12,908 lightning recordings, the classification process successfully identified 550 data points as Negative Narrow Bipolar Events (NNBE) and 2,314 as Positive Narrow Bipolar Events (PNBE), with the remaining 10,044 categorized as Non-NBE. Characteristic analysis revealed that NNBE had an average pulse duration of 32.43 ± 9.54 µs and a rise time of 4.97 ± 4.65 µs. Meanwhile, PNBE exhibited an average pulse duration of 35.1 ± 10.72 µs with a rise time of 5.08 ± 6 µs. Model validation using a confusion matrix showed an overall accuracy of 96%, with a precision of 93% and a recall of 98% for the NNBE class, and a recall of 91% for the PNBE class. These results indicate that the KNN method is effective for NBE lightning classification, despite a slight weakness in distinguishing between the PNBE and Non-NBE classes. This study contributes to the automation of lightning classification using a machine learning approach and provides a foundation for developing more precise extreme weather detection systems.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2507004383	T179672	T1796722025	Central Library (Reference)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1796722025

Publisher

Indralaya : Prodi Teknik Elektro, Fakultas Teknik Universitas Sriwijaya., 2025

Collation

xxiii, 443 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

621.310 7

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Teknik elektro

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment