Determination of perspective towards humanoid robots in accommodation sector with text mining

With the development of robot technology, many types of robots have seen in business and daily life. Humanoid robots are one of these robots. Humanoid robots, with their humanoid look, are mostly used in business areas that require human interaction. This study is examined customer perspective on humanoid robots that work in accomodation industry. Using the purposive sampling technique the sample of study was determined as the customers of Henn-na Hotel. In the study, data set was created for Henn-na Hotel by taking into account the online customer comments made on the Booking and Agoda platforms between June 2018 and October 2021. After the data was preprocessed, analyses were conducted using natural language procossing methods. It was divided into three categories: positive, negative, and neutral. The number of positive comments was 1314, the number of negative comments was 305, and the number of neutral comments was 241 as a result of sentiment analysis. The Python programming language was used for relevant analysis. Machine learning algorithms were also utilized to assess the success of sentiment classification as a result of online comment sentiment analysis. The Orange software was used to run machine learning applications. While the success of the algorithms was seen with the highest 20 folds, Logistic Regression and Neural Network were the most successful algorithms. Finally, the word2vec approach was used to examine online comments in greater depth. Analyses were carried out on the words within the scope of the research topic using the skip-gram and CBOW models, which were based on this approach. These models, which highlight word similarities, show that positive terms like friendly, enjoyable, and beautiful are used for robots in the comments; phrases like difficult, compelling, and usefulness reveal that customers have problems utilizing robots and complain about them.