Machine Learning for Forecasting Future Reservations’ Ratings : Radisson Blu Seaside in Helsinki
Nie, Tong (2023)
Nie, Tong
2023
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-fe2023050340807
https://urn.fi/URN:NBN:fi-fe2023050340807
Tiivistelmä
In the current age of internet and big data, it is imperative for hotels to enhance their online reputation to remain competitive and profitable. This research presents a new perspective on how hotels can maintain and improve their online reputation through the use of machine learning techniques to predict the ratings of reservations. The approach involves analysing data that customers provide when booking a room. Additionally, the study explores how insights gleaned from online textual reviews can be used by hotel managers to address negative ratings.
The study's primary objective is to assess the effectiveness of machine learning in predicting negative instances, a critical factor in managing online reputation. The best performing models achieved a 60% accuracy in classifying negative instances. However, increasing the number of predicted true negative instances also increased the number of false negative instances. This result was primarily due to the unpredictability of customer behaviour, making it difficult to accurately predict ratings.
Despite not achieving the desired result, this study presents a novel direction for future research and provides suggestions for future research ideas. By utilizing machine learning algorithms to analyse customer data, hotels can better understand their customer's preferences, allowing them to improve their online reputation and ultimately improve their bottom line.
The study's primary objective is to assess the effectiveness of machine learning in predicting negative instances, a critical factor in managing online reputation. The best performing models achieved a 60% accuracy in classifying negative instances. However, increasing the number of predicted true negative instances also increased the number of false negative instances. This result was primarily due to the unpredictability of customer behaviour, making it difficult to accurately predict ratings.
Despite not achieving the desired result, this study presents a novel direction for future research and provides suggestions for future research ideas. By utilizing machine learning algorithms to analyse customer data, hotels can better understand their customer's preferences, allowing them to improve their online reputation and ultimately improve their bottom line.
Kokoelmat
- 512 Liiketaloustiede [433]