Penemuan Pola Penyebab Kecelakaan Pesawat Menggunakan Metode Targeted Balanced Utility Pattern Mining

Alviando, Rizky (2025) Penemuan Pola Penyebab Kecelakaan Pesawat Menggunakan Metode Targeted Balanced Utility Pattern Mining. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5002211001-Undergraduate_Thesis.pdf - Accepted Version Restricted to Registered users only Download (2MB) | Request a copy

Abstract

Informasi terkait kecelakaan pesawat dikumpulkan dalam waktu yang cukup lama dan prosedur yang rumit. Untuk meningkatkan tingkat keselamatan penerbangan global, fokus analisis investigasi masih dilakukan secara manual dan sulit sejauh ini. Penelitian Tugas Akhir ini bertujuan untuk mengekstraksi informasi data penyebab kecelakaan pesawat berdasarkan pendekatan data mining yang dapat digunakan untuk membantu proses analisisnya. Jika dikaitkan dengan metode data mining, studi terkini mencari pola targeted utility (TU) yang berfokus pada pola item spesifik yang paling berisiko. Pada permasalahan penelitian Tugas Akhir ini, pola TU masih kurang representatif dalam menjelaskan fitur penyebab kecelakaan pesawat jika terlalu berfokus pada risiko salah satu kejadian atau outlier. Oleh karena itu, penelitian Tugas Akhir ini mengusulkan sebuah metode Targeted Balanced Utility Pattern Mining atau T-BUM yang berusaha untuk mengekstraksi pola penyebab kecelakaan pesawat berdasarkan kejadian yang tidak hanya memilki risiko tinggi, tetapi juga unggul secara statistik. Secara umum, T-BUM didasarkan pada sebuah trade-off dari parameter utility dan support. T-BUM diuji pada data publik kecelakaan pesawat Aircraft Accident Dataset. Dari hasil eksperimen yang dilakukan, T-BUM berhasil mengekstrak pola penyebab kecelakaan dengan lebih efisien dan representatif dibandingkan model baselinenya.
====================================================================================================================================
Information related to aircraft accidents is collected over a long period and through complex procedures. To enhance global aviation safety, the focus of investigative analysis remains largely manual and challenging to this day. This thesis aims to extract causal patterns of aircraft accidents using a data mining approach to support and improve the analysis process. In the context of data mining, recent studies have explored targeted utility (TU) patterns, which emphasize high-risk specific itemsets. However, in this study, TU patterns are considered less representative in explaining the features of accident causes, especially when they overly focus on a single high-risk event or outlier. Therefore, this thesis proposes a method, Targeted Balanced Utility Pattern Mining or T-BUM, designed to extract causal patterns of aircraft accidents by considering not only high-risk events but also those that are statistically significant. In general, T-BUM is based on a trade-off between utility and support parameters. The method is evaluated on the publicly available Aircraft Accident Dataset. Experimental results demonstrate that T-BUM can extract causal patterns more efficiently and representatively compared to its baseline model.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Accident Rate, Balanced Utility, Kecelakaan Pesawat, Penyebab Kecelakaan, Targeted Utility Mining, Aircraft Accident, Accident Causes
Subjects:	Q Science > QA Mathematics Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) Q Science > QA Mathematics > QA9.58 Algorithms Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics) T Technology > T Technology (General) > T55.3.H3 Hazardous substances--Safety measures. T Technology > T Technology (General) > T55 Industrial Safety
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Rizky Alviando
Date Deposited:	01 Aug 2025 02:27
Last Modified:	01 Aug 2025 02:27
URI:	http://repository.its.ac.id/id/eprint/125645

Actions (login required)

View Item