AI Driven Tourism Forecasting - Case Lenoni Oy
Bukhari, Aneeq (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025120131253
https://urn.fi/URN:NBN:fi:amk-2025120131253
Tiivistelmä
The unpredictability of tourism causes some difficulty for tourism service providers to accurately forecast the demand and therefore make informed decisions. This thesis explored methods that could be employed with Artificial Intelligence to improve demand forecasting, and provide additional insight into customer behaviour for Helsinki Tours, which is a division of Lenoni Oy.
The aim of this thesis was to explore and assess the effectiveness of artificial intelligence (AI)-based methods for developing an improved model to describe how tourists respond to changes in their environment, such as seasonality ornatural disasters, etc., through time series forecasting using long short-termmemory (LSTM) methods. In addition, this research also sought to assess the sensitivities associated with using sentiment analysis techniques on unstructured data collected through online reviews & social media platforms.
Responses were collected from 12 respondents, who participated in semi- structured interviews and filled out Google Forms surveys after the interviews that were identical to Interviews. All twelve respondents reported common operational challenges for tourism service providers, including late cancellations, increased demands placed on staff, and poor communication between staff and customers. Of the twelve respondents surveyed, 83% already have employed some form of artificial intelligence techniques and view AI as an asset, while 16.7% currently do not use AI techniques and expressed concerns about implementing AI technology.
The main results revealed that managers anticipate a growing trend of support for the adoption of AI by their organizations.Moreover, the results indicate that using LSTM (long short-term memory) modeling and sentiment analysis have a better forecasting performance than the prediction models currently used by Lenoni Oy.
The aim of this thesis was to explore and assess the effectiveness of artificial intelligence (AI)-based methods for developing an improved model to describe how tourists respond to changes in their environment, such as seasonality ornatural disasters, etc., through time series forecasting using long short-termmemory (LSTM) methods. In addition, this research also sought to assess the sensitivities associated with using sentiment analysis techniques on unstructured data collected through online reviews & social media platforms.
Responses were collected from 12 respondents, who participated in semi- structured interviews and filled out Google Forms surveys after the interviews that were identical to Interviews. All twelve respondents reported common operational challenges for tourism service providers, including late cancellations, increased demands placed on staff, and poor communication between staff and customers. Of the twelve respondents surveyed, 83% already have employed some form of artificial intelligence techniques and view AI as an asset, while 16.7% currently do not use AI techniques and expressed concerns about implementing AI technology.
The main results revealed that managers anticipate a growing trend of support for the adoption of AI by their organizations.Moreover, the results indicate that using LSTM (long short-term memory) modeling and sentiment analysis have a better forecasting performance than the prediction models currently used by Lenoni Oy.