Enhancing Flight Delay Prediction through Feature Engineering in Machine Learning Classifiers - A Real-Time Data Streams Case Study

Aug 5, 2026

Upload by Logic

Feature engineering creates and selects informative features from raw data to enhance machine learning model accuracy, especially for real-time streams whose distributions shift continuously. This paper presents a flight information system case study where engineered time-, trend-, and error-based features improve classifiers for predicting flight delays. It applies CTAO preprocessing, then SCSO feature extraction and Enhanced harmony search optimization. The resulting feature set selects the nine most relevant signals, outperforming raw-feature baselines for real-time delay prediction.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Introduction

Feature engineering for real-time data streams

Feature selection: filter vs wrapper models

Related Works

Feature engineering with genetic algorithms and hybrid learning

Machine learning plus feature engineering for text classification

Abstract—The process of creating and selecting features from raw data to enhance the accuracy of machine learning models is referred to as feature engineering. In the context of real-time data streams, feature engineering becomes particularly important because the data is constantly changing and the model must be able to adapt quickly. A case study of using feature engineering in a flight information system is described in this paper. We used feature engineering to improve the performance of machine learning classifiers for predicting flight delays and describe various techniques for extracting and constructing features from the raw data, including time-based features, trend-based features, and errorbased features. Before applying these techniques, we applied feature pre-processing techniques, including the CTAO algorithm for feature preprocessing, followed by the SCSO (Sand cat swarm optimization) algorithm for feature extraction and the Enhanced harmony search for feature optimization. The resultant feature set contained the 9 most relevant features for deciding whether a flight would be delayed or not. Additionally, we evaluate the performance of various classifiers using these engineered features and contrast the results with those obtained using raw features. The results show that feature engineering significantly improves the performance of the classifiers and allows for more accurate prediction of flight delays in real-time. Keywords-Feature Engineering, Machine Learning, Classifiers, Real-time Data Streams, Flight Information System. Enhancing Flight Delay Prediction through Feature Engineering in Machine Learning Classifiers: A Real Time Data Streams Case Study Ms. Shailaja B. Jadhav1, Dr. D. V. Kodavade2 1Assistant Professor : Dept. of Computer Engg., Marathwada Mitramandal ‘s College ofEngg., Pune Research Scholar – Department of Technology, Shivaji University, Kolhapur Maharashtra-India [msgshalom@gmail.com](msgshalom@gmail.com) 2Professor, Dept. OfCSE and IT DKTE’s Institute of Engg. and Technology Ichalkaranji, Kolhapur-India [dvkodavade@gmail.com](dvkodavade@gmail.com) I. INTRODUCTION Creating new features from raw data is a vital aspect in the machine learning workflow, by this way the performance of the models can be optimized. In the context of real-time data streams, where the data is constantly changing and the model must be able to adapt quickly, feature engineering becomes especially important. Recent research has focused on various techniques for extracting and constructing features from raw data, including time-based features, trend-based features, and error-based features [1] [2] [3] . These techniques have been applied to a variety of real-time data streams, including flight information systems, financial markets, and social media networks [4] [5] [6] . The use of feature engineering in machine learning classifiers for real-time data streams has been shown to significantly improve the performance of the classifiers and enable more accurate prediction of events such as flight delays and stock price movements [4] [5]. For example, in a study of a flight information system, the authors used feature engineering to improve the performance of machine learning classifiers for predicting flight delays [4] . Feature engineering is a critical step in machine learning, it is the process of creating new information from raw data that can be used to train a model [7] . A variety of techniques can be used for feature engineering, such as time-based, trend-based and error-based. In [8] it was demonstrated that by using engineered features, the performance of classifiers was significantly improved and the prediction of flight delays in real-time was more accurate when compared to results obtained from raw features only. It is common practice in streaming data to first optimize and engineer features before utilizing data mining algorithms as it can lead to better performance. In the

FAQs

What is the role of feature engineering in real-time data stream prediction?

Feature engineering transforms raw, continuously changing data into informative inputs so models can adapt quickly and make more accurate predictions.

Which feature engineering techniques are applied before training classifiers?

The paper uses CTAO for feature preprocessing, then SCSO for feature extraction, and Enhanced harmony search for feature optimization.

How do the engineered features compare with raw features for flight delay prediction?

Engineered features significantly improve classifier performance and yield more accurate real-time flight delay predictions than raw features alone.

Aug 5, 2026

Upload by Logic