Studying the Invasion of Drones in Indigenous Areas Using Machine Learning Techniques

Authors

  • Samra Nawazish Software Engineering Department, University of Engineering and Technology, Taxila, Pakistan.
  • Amin Ullah Postdoc Research Associate, School of Computer Science & Engineering, Southeast University, Nanjing, Jiangsu China.
  • Jawad Hassan Deparmtment of Computer Science and Information Technology, University of Lahore
  • Saddam Hussain Khan Department of Computer Systems Engineering, University of Engineering and Applied Science, Swat Pakistan
  • Abdul Aziz Phd Scholar, School of Computer Science and Engineering, Southeast University, Nanjing, China.
  • Riaz Ud Din Department of Computer Systems Engineering, University of Engineering and Applied Science, Swat, Pakistan.
  • Muhammad Adeel Department of Computer Science, National Textile University, Faisalabad Pakistan

Keywords:

Drone attack, civilian, Data Mining, Pattern Analysis

Abstract

The prevalence of drones in contemporary times has become widespread, representing a pivotal technological advancement in aviation—characterized by its autonomous, pilot-less nature. However, its predominant application for surveillance and targeted operations, particularly by the United States of America (USA), has sparked vehement criticism due to perceived violations of human rights on a global scale. Especially, Pakistan-like countries have borne the brunt of drone strikes, with the Federally Administered Tribal Areas being the primary target, accounting for over 90% of these attacks. This research delves into the profound impact of drone strikes, focusing on the often-overlooked innocent victims, including women and children, as well as the consequential damage to the affected regions. In this paper, we posit that a classification-based approach offers a more comprehensive and statistically informative means of elucidating patterns inherent in the data. By doing so, we aim to shed light on the effectiveness of targeted killings in the context of counter-terrorism. The proposed approach includes machine learning algorithms, such as ZeroR, J48, Naive Bayes, and OneR that have been employed to meticulously analyze the dataset and unveil hidden patterns. In particular, the J48 algorithm demonstrated exceptional performance, accurately discerning casualties within the standard Kaggle noisy dataset. The Weka tool, known for its advanced capabilities, played a pivotal role in this analysis, handling crucial tasks such as initial pre-processing, numeric to nominal conversion, and replacing missing values. This integrated approach ensures a robust exploration of the dataset, leveraging the strengths of diverse algorithms and sophisticated tools for comprehensive insights. This departure from traditional legal analyses broadens the discourse surrounding drone warfare, emphasizing the importance of data-driven insights in understanding the broader implications of these operations.

Downloads

Published

2024-03-18

How to Cite

[1]
S. . Nawazish, “Studying the Invasion of Drones in Indigenous Areas Using Machine Learning Techniques”, UCP-JEIT, vol. 1, no. 2, pp. 17–23, Mar. 2024.