Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций
Конференция: Computational Methods in Systems and Software 2023 (CoMeSySo2023); Zlín, Czech Republic; Zlín, Czech Republic
Год издания: 2024
Ключевые слова: data set analysis, airline passenger satisfaction, correlation analysis, neural network prediction, decision tree algorithm
Аннотация: This paper presents an analysis of a data set to determine the factors influencing airline passenger satisfaction. The study examines various criteria such as gender, customer type, age, travel type, class, and range to assess their impact on passenger satisfaction. The dataset consists of 25 columns, including attributes like Wi-FПоказать полностьюi availability, convenience of online booking, seat comfort, in-flight entertainment, baggage handling, and overall satisfaction. The sample is relatively balanced, with equal representation of men and women, predominantly repeat customers, and a majority flying for business purposes. Key findings include a strong correlation between departure and arrival delays, higher satisfaction among passengers in business class, and positive ratings for Wi-Fi service correlating with overall satisfaction. Correlation analysis reveals interdependencies between different attributes, such as the influence of cleanliness on seat comfort and food and beverage ratings. In addition, a neural network forecasting model is used to estimate the average ratings of passengers, although with low accuracy, which was later excluded. Finally, a decision tree algorithm is utilized to identify the most significant attributes affecting passenger satisfaction words.
Журнал: Data Analytics in System Engineering
Выпуск журнала: 910-3
Номера страниц: 434-458
Место издания: Springer Nature Switzerland