Airline Passenger Feedback Data

Columns

• title: The title or header given to the review [2].
• rating: The customer's satisfaction rating, provided on a scale out of 10 [2]. The distribution of ratings shows a concentration at lower scores, with 352 reviews between 1.00-1.90 and 113 reviews between 9.10-10.00 [2, 3].
• review_date: The specific date on which the review was submitted [2].
• review: The actual text content of the customer's review [2].
• trip_verified: A boolean value indicating whether the reported trip was verified [2]. 943 (79%) of the trips are verified, while 257 (21%) are not [3].
• aircraft: Specifies the type of aircraft flown on [2]. The A320 accounts for 40% of mentions, with 577 other unique aircraft types categorised as 'Other', making up 48% of the data [3].
• type_of_traveller: Describes the nature of the traveller, such as 'Solo Leisure' or 'Couple Leisure' [2]. Both 'Solo Leisure' and 'Couple Leisure' each account for 32% of the traveller types, with 428 other unique types making up the remaining 36% [3].
• seat_type: Details the type of seat or class flown in [2]. 'Economy Class' is the most common at 54%, followed by 'Business Class' at 33%, with 155 other unique seat types representing 13% [3].
• route: The specific flight route, indicating the origin and destination [2]. There are 808 unique routes recorded in the dataset [3].
• date_flown: The date on which the flight actually took place [2]. This column contains 768 unique date values [3].

Distribution

Usage

• Modelling: Developing predictive models, for example, to forecast customer satisfaction or identify factors influencing review scores [1].
• Predictive Analysis: Analysing trends and predicting future customer sentiment or operational issues [1].
• Feature Engineering: Extracting meaningful features from unstructured text reviews and other categorical data for machine learning tasks [1].
• Exploratory Data Analysis (EDA): Gaining insights into customer feedback patterns, common complaints, service quality, and performance trends over time [1].
• Natural Language Processing (NLP): Performing sentiment analysis, topic modelling, or text classification on the review content [1].

Coverage

• Geographic Scope: The dataset's coverage is global [5].
• Time Range: The date_flown column covers flights from 17th January 2022 to 21st December 2022 [6]. Review dates also fall within this period, providing a snapshot of customer feedback for this timeframe.
• Demographic Scope: It includes various traveller types, such as solo leisure and couple leisure, which can offer insights into different customer segments [3].

License

Who Can Use It

• Data Scientists: For building and testing machine learning models related to customer satisfaction and service quality.
• Business Analysts: To understand customer feedback, identify pain points, and inform business decisions aimed at improving service.
• Aviation Industry Researchers: For academic or industry-specific studies on airline performance and customer experience.
• Developers: To create applications that leverage customer insights or integrate review data for reporting.

Dataset Name Suggestions

• British Airways 2022 Customer Reviews
• Airline Passenger Feedback Data
• BA Customer Experience Dataset
• Global Airline Review Archive

Attributes