Search results

Budget hotels are becoming an emerging industry for convenience and affordability, where consumer sentiments are of paramount importance. Tourism has become increasingly dependent on social media and online platforms to gather travel-related information, purchase travel products, food, lodging, etc., and share views and experiences. The user-generated data helps companies make informed decisions through predictive and behavioural analytics.

Design/Methodology/Approach: This study uses text mining, deep learning, and machine learning techniques for data collection and sentiment analysis based on 117,151 online reviews of the customers posted on the TripAdvisor website from May 2004 to May 2019 from 197 hotels of five prominent budget hotel groups spread across India using Feedforward Neural Network along with Keras package and Softmax activation function.

Findings: The word-of-mouth turns into electronic word-of-mouth through social networking sites, with easy access to information that enables customers to pick a budget hotel. We identified 20 widely used words that most customers use in their reviews, which can help managers optimise operational efficiency by boosting consumer acceptability, satisfaction, positive experiences, and overcoming negative consumer perceptions.

Practical Implications: The analysis of the review patterns is based on real-time data, which is helpful to understand the customer’s requirements, particularly for budget hotels.

Originality/Value: We analysed TripAdvisor reviews posted over the last 16 years, excluding the Corona period due to industry crises. The findings reverberate in consonance with the performance improvement theory, which states feed-forward a neural network enhances organisational, process, and individual-level performance in the hospitality industry based on customer reviews.

One of the most contested and anticipated research issues is the acceptability of using recycled aggregates instead of fresh aggregates. This study aims to look at the possibility of replacing fresh aggregates with 15%, 30%, 60% and 100% recycled aggregates.

The research is divided into two stages. The compressive, split tensile, flexural and bond strength of the various mixes were examined in the first phase using untreated recycled concrete aggregates (RCA). The second phase entails chemically treating RCA with a 10% 0.1 M sodium metasilicate solution to evaluate differences in strength, indicating the success of the treatment performed. Microstructural experiments such as scanning electron microscopy and X-ray diffraction were also conducted to evaluate the formation of interfacial transition zone (ITZ) in treated and untreated RCA specimens.

The observed findings reveal a decrease in concrete strength with increasing RCA concentration; however, when treated RCA was used, the strengths increased significantly when compared to untreated samples. The findings also include curves indicating the correlation between compressive strength and other mechanical strength parameters for an optimum mix of concrete prepared with 30% RCA replacement.

The study through its novel approach, demonstrates the effect of pretreatment of RCA in the absence of any standardized chemical treatment methodology and presents significant potential in minimizing reliance on fresh aggregates used in concrete, lowering building costs and promoting the use of waste materials in construction.