Analysis of Online Hotel Reviews During the COVID-19 Pandemic Using Topic Modeling

Analysis of Online Hotel Reviews During the COVID-19 Pandemic Using Topic Modeling

Özlem Ergüt
DOI: 10.4018/978-1-7998-8231-2.ch023
(Individual Chapters)
No Current Special Offers


The world is facing the COVID-19 pandemic that has impacted economies and millions of people worldwide. The fact that COVID-19 is highly contagious from person to person has greatly affected the daily lives of people, and it has also had a devastating effect on many sectors, particularly the tourism industry. In order to mitigate losses for the tourism sector and for it to gain a new dynamism under the current pandemic conditions, monitoring and analyzing online reviews is an important factor for better understanding the needs and desires of customers. The purpose of this study was to determine the main topics in online reviews by foreign guests staying in İstanbul during the pandemic period using text mining techniques. The information obtained as a result of the analysis is important in terms of understanding how to manage the current situation, developing suggestions for solutions, improving service quality, making future decisions, and adapting to the new normal.
Chapter Preview


COVID-19, which started and emerged in one of China's most populous cities, Wuhan, spread rapidly as a result of being highly contagious from person to person and affected millions of people worldwide. Many measures and practices have been put into effect to reduce interpersonal contact in order to slow down the spread of COVID-19. As a result of the restrictions that occurred across the country of Turkey, restaurants, shopping centers were closed, markets introduced limited hours, and curfews were started where people had to spend most of their time at their homes. The rapid spread of COVID-19 from person to person has not only been within individual countries but also has brought about travel restrictions between countries.

COVID-19, which has had a great impact on economic activity due to the ease of spread, severity and mortality rate of the disease, uncertainty of appropriate policies, and individual behavior (Song, Yeon, & Lee, 2021). The tourism sector was among the sectors most affected as a result of the stopping flights between countries, the closure of hotels, and travel restrictions. The tourists’ fears about travel itself as well as questions and uncertainty about conditions in the destination country and its inability to manage risk have emerged as major obstacles to people traveling. According to United Nations World Tourism Organization (UNWTO) World Tourism Barometer, international tourist arrivals fell by 72% in January-October 2020 over the same period last year, and the decline in tourist arrivals in 2020 is equivalent to a loss of about one billion arrivals and 1.1 trillion dollars in international tourism receipts (UNWTO, 2020).

Travel restrictions, one of the decisions taken within the scope of COVID-19 measures, made tourists feel anxious about travel. The most important issue in deciding on the travel after the pandemic is safety, and hoteliers should pay attention to this issue. Customers are concerned about issues such as how safe the place they stay in including whether the rooms are clean, the hygiene of the staff, and the layout of the dining areas. In many areas, from transport to accommodation, in order to restore customer confidence, a number of programs have been put into place in many countries including Turkey where the Safe Tourism Certification Program, one of the first examples in the world, was put into use. With this certificate, the people who want to vacation in Turkey may have confidence in a wide range of services from transportation to accommodation and facilities’ professionals accommodating a wide range of health conditions. Under the conditions of this certificate, accommodation, food and beverage facilities, tour and transfer vehicles, congress and art facilities, theme parks, mechanical lines, and sea tourism are audited at international standards within the framework of separate criteria, and reports are issued as a result of the conformity assessment (TGA, 2020).

Key Terms in this Chapter

Topic Modeling: Topic modeling is an unsupervised machine learning technique that finds natural sets of items and similar expressions by identifying the words that best represent a set of documents.

Customer Satisfaction: It is the level of satisfaction with the services received after supplying the needs of the customer.

Latent Dirichlet Allocation: It is one of the most used topic modeling methods used to discover latent semantic relationships in large collections of text documents.

Unstructured Data: It is data that has no predefined data model or structure, such as image files, text files such as pdf, word, and text.

Text Mining: Text mining is the process of identifying information in unstructured data, searching, and identifying hidden and different patterns.

COVID-19: It is a disease caused by a new strain of coronavirus.

Online Reviews: It is a review of a product or service where it reflects the opinions and experiences of a customer purchasing a product or service.

Complete Chapter List

Search this Book: