Data Mining for Hospitality Industry: A Humanizing Approach.

摘要

In the age of big data, the amount of accessible information is growing rapidly, and business world is becoming increasingly data-driven. Especially in marketing, the challenges and changes associated with big data are dramatic. Extracting actionable knowledge from vast, complex datasets is simply impossible using traditional data processing and analysis applications and methods. Thus, marketers must embrace new technologies and methods to overcome these challenges and better understand their customers.;As a new technology and analysis method, data mining helps marketers harness the power of big data in order to correctly identify customer preferences and behaviors, ensure personalized and consistent service, and develop products and services that are most appealing to the market. Given the vast amount of data that exists, useful knowledge extraction can be challenging. Data mining reveals rules and/or patterns that are undetected by traditional methods. Therefore, data mining techniques have tremendous potential to solve real problems in the business world that were not able to be addressed previously. For example, by using data mining techniques, members of the food-service industry could identify frequently co-purchased items within one category or across categories.;Despite their great potential, many data mining techniques have yet to be implemented. Integrating new techniques into business applications is difficult and challenging. The use of data mining techniques remains especially limited among hospitality and tourism researchers, despite the developmental maturity of such techniques in the computer science field. Why is this true? To put it simply, hospitality and tourism management researchers tend to have humanized mindsets, whereas computer scientists tend to have computerized mindsets. Although hospitality marketers do not lack access to data or computerized systems, they tend to use only basic system functions to generate transaction summaries, and most do not understand how to fully utilize valuable data. One way to bridge the gap between the computerized mindset reflected in data mining techniques and the humanized mindset required to address actual consumer demands in the hospitality industry is to represent knowledge extracted from huge datasets in the form of human language.;For hotels, restaurants, travel agencies, and airline companies, all customer purchase transactions are stored in point of sale (POS) systems. POS data are at the core of the hospitality industry. In this research, I transform a large POS dataset (more than 40,000 rows) into meaningful, humanized knowledge (text and pictures) that are easy to mentally process and understand. The generated outcomes can be understood by users regardless of their areas of expertise. The findings generated by the proposed methods will help marketers understand their customers' behaviors and guide the development of marketing campaigns.;Key words: data mining, hospitality industry, point-of-sale (POS) data, association analysis.