Work

An Aspect based Sentiment Analysis of Tour and Travel Recommendation Approach using Machine Learning

Year: 2023
Type: article
Abstract: Consumers nowadays are more inclined to depend on internet aspect to help them make educated selections when purchasing products and services. These technologies enable businesses to get insight from the experiences of their consumers and pinpoint areas in which they may enhance the products and services they provide. A survey found that 82 percent of adults in the United States have relied on online reviews when making a purchase. Approximately forty percent of them report that they have told other individuals about the things that they have purchased. Regrettably, it is exceedingly difficult to make accurate predictions about the ratings that new users and goods will get in the recommender systems. The difficulty in question is often known as the cold-start problem. In this article, we will discuss a new method of filtering known as hybrid filtering, which combines collaborative filtering, content-based filtering, and demographic filtering. The hybrid filtering approach that has been presented takes into consideration the numerous demographic particulars of a user in order to forecast the ratings and locate other items in the area that are comparable. This strategy gets beyond the problems that are inherent in more conventional techniques of suggestion, such as CF and CB. After that, the points of interest (POIs) that are pertinent to the user are extracted using the data that were gathered for this article. With the help of the data that was obtained, we were able to carry out a prediction study on the ratings for the various airline services. The findings of this study showed that the most common complaints about business class were related to the quality of the food and the friendliness of the staff, whereas the most common complaints about economy class were related to the level of comfort provided by the seats and the amount of legroom available. In this research, the machine learning (ML)-based hybrid filtering algorithm that was suggested worked quite well. It has the potential to assist in the resolution of the cold-start issue by determining the goods that are most likely to be valuable to the customers.
Language: English
Open Access status: diamond
OpenAlex ID: W7217595648
FWCI: 0.408
Cites: 28
Cited by: 3
Related to: 10