Print

WP2: Tool Development - Leader: Mikkel N. Schmidt

Members: Kristoffer Jon Albers, Postdoc (DTU) Marisciel Palima, Research Assistant (CBS) Morten Mørup, Associate Professor (DTU)  Fumiko Kano Glücksrad, Associate Professor (CBS) Mikkel N. Schmidt, Associate Professor (DTU) Executive summary of the activities completed by WP2   The WP2 team has developed a data analytic framework that integrates the state-of-the-art segmentation algorithm called Stochastic Block Model (SBM) based on the Bayesian statistics (Schmidt & Mørup, 2013; Herlau et al. 2012; Glückstad et al. 2014). This segmentation algorithm groups people who share similar response patterns to specific survey questions. In the UMAMI project, we have segmented people according to their responses to a so-called Portrait Value Questions (PVQ) that consist of 21 items representing Schwartz Theory of Ten Basic Human Values (Schwartz, 2006), the scale acknowledged by psychologists, sociologists marketing scientists, but also widely used by marketing practitioners. The performance and usability of the PVQ questions and the segmentation algorithm were verified by classifying more than 35000 respondents from a multinational database, "European Social Survey (ESS)". The reliability of segments extracted from the 21 EU countries and their usability to predict external variables (attitudes to immigrants, attitudes to European unification, satisfaction in life etc.) were also tested with a “between-segment” analysis method called “fuzzy-set Qualitative Comparative Analysis (fsQCA). After the applicability of the Schwartz theory, the segmentation method, and the pre- & post-processing methods were validated with the ESS data, the same methodological framework was applied to classify 6000 respondents collected from the main Chinese survey into 22 segments, and characteristics of the respective segments (various independent variables and their relation to a behavioral intention, i.e. willingness to visit a destination) were compared based on the aforementioned fsQCA method. In addition, the team has developed a visualization framework of segmentation results by integrating “Means-End-Chain” theory (Gutman, 1982) widely used by both academics and practitioners to develop a positioning strategy of products. This visualization framework enables users to analyze relations between personal values, psychological and functional motivations and product attributes, i.e. in case of the tourism product, descriptive destination attributes.   In Dec. 2018, the development of the prototype of UMAMI data analytic framework has been completed and UMAMI's stakeholders are now able to receive a prototype software and an instruction guide upon request for the purpose of the internal assessment and explore data collected from China as part of WP3. Methods employed as part of the data analytic framework have been published in the proceedings of several academic conferences and some of them were selected as “best excellent papers” in the major Marketing and Tourism research conferences:   Methods employed as part of the data analytic framework have been introduced in several conferences in below:  
  • Fumiko Kano Glückstad; Mikkel N. Schmidt; Morten Mørup / Testing a Model of Destination Image Formation : Application of Nonparametric Bayesian Relational Modeling to Destination Image Analysis. In: 2018 Global Marketing Conference at Tokyo Proceedings. ed. /Jeonghye Choi. Seoul : Global Alliance of Marketing & Management Associations 2018, p. 63-64 (Global Marketing Conference Proceedings) BEST EXCELLENT PAPER AWARD
  • Kristoffer Jon Albers; Mikkel N. Schmidt; Marisciel Litong-Palima; Morten Mørup; Rasmus Bonnevie; Fumiko Kano Glückstad / Understanding Mindsets Across Markets, Internationally : A Public-private Innovation Project for Developing a Tourist Data Analytic Platform. In: Proceedings of the 42nd IEEE Annual Computer Software and Applications Conference: COMPSAC 2018. Volume 2. . ed. /Sorel Reisman; Sheikh Iqbal Ahamed; Claudio Demartini; Thomas Conte; William Claycomb; Motonori Nakamura; Edmundo Tovar; Stelvio Cimato; Chung-Horng Lung; Hiroki Takakura; Ji-Jiang Yang; Toyokazu Akiyama; Zhiyong Zhang; Kamrul Hasan. Los Alamos, CA : IEEE 2018, p. 159-164
 
IFD_logo
UMAMI is the interdisciplinary research project funded by Innovation Fund Denmark for the period 2017-2020.



Last updated by Fumiko Kano Glückstad 23/07/2019