Attributes prediction from IoT consumer reviews in the hotel sectors using conventional neural network: deep learning techniques

The Internet of Things (IoT) plays an important role in helping the hotel industry increase customer satisfaction while maintaining affordable costs. IoT consumers review and rate the hotels online. The ratings are based on the Value, Apartment, Site, Sanitation, Front Desk, Facility, Professional Provision, Internet, and Packing. Traditional systems that predict hotel ratings with minimum accuracy create complexity through their analysis of the ratings. Thus, the effective deep learning techniques are used to analyze the reviews in order to help consumers choose better hotels. In this paper, different classification algorithms, such as convolutional neural network-based deep learning (CNN-DL), support vector machine network-based deep learning are applied to predict attributes. The system utilizes the TripAdvisor site, which is a well-known America dataset for examining system efficiency. The experimental results show that the CNN-DL algorithm has better classification accuracy and a lower error rate as compared to other algorithms.