AUTHOR=Cao Enguo , Jiang Jinzhi , Duan Yanjun , Peng Hui 

TITLE=A Data-Driven Expectation Prediction Framework Based on Social Exchange Theory

JOURNAL=Frontiers in Psychology

VOLUME=Volume 12 - 2021

YEAR=2022

URL=https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.783116

DOI=10.3389/fpsyg.2021.783116

ISSN=1664-1078

ABSTRACT=Along with the rapid application of new information technologies, the data-driven era is coming, and 11 online consumption platforms are booming. However, massive user data has not been fully 12 developed for design value, and the application of data-driven methods of requirement engineering 13 needs to be further expanded. This paper proposes a data-driven expectation prediction framework 14 based on social exchange theory, which analyzes user expectations in the consumption process, and 15 predicts improvement plans to assist designers make better design improvement. According to the 16 classification and concept definition of social exchange resources, consumption exchange elements 17 were divided into seven categories: money, commodity, services, information, value, emotion and 18 status, based on these categories, two data-driven methods: word frequency statistics and scale 19 surveys, were combined to analyze user generated data. Then mathematical expectation formula was 20 used to expand user expectation prediction. Moreover, by calculating mathematical expectation, 21 explicit and implicit expectations are distinguished to derive a reliable design improvement plan. To 22 validate its feasibility and advantages, an illustrative example of CoCo Fresh Tea & Juice service 23 system improvement design is further adopted. As an exploratory study, it is hoped that this work 24 provides useful insights into the data mining process of consumption comment.