With the rapid development of social networking sites, such as Facebook, Twitter, and Pinterest, both consumers and companies are paying increased attention to social commerce. It is characterized by massive and unstructured content generated along with users’ sharing, commenting, and other interaction behaviors.

Compared with the traditional data sources, such as surveys, experiments, and field experiments, which are costly in terms of time and money, user-generated content (UGC) on social commerce platforms provides more reliable, more accessible, and more informative raw materials for both academics and industrials to excavate consumers’ requirements and preferences, understand their underlying behaviors, and therefore, support scientific decision-making, including but not limited to marketing, product design, operations, and business strategies decisions. Moreover, the emergence of social commerce inevitably reshapes our behaviors, and even shifts our behavior patterns, and changes the way we think. Therefore, classic theories and models proposed in previous research should be testified in this newly emerging environment. Furthermore, probably, by fully exploiting UGC, some new phenomena can be discovered and based on these findings, establishing new theories which can contribute to explaining consumers’ psychologies and behaviors.

While UGC mining has enormous value, however, it is of huge difficulties to extract valuable knowledge from the massive, various, but with low-value density and high noise UGC. We should notice that innovations in data science techniques are the critical key to solve these problems. As an interdisciplinary science, it is highly related to mathematics, statistics, and computer science including machine learning, database, and pattern recognition. Many techniques in these field has been attempt to utilize for mining data, for example, natural language processing (NLP) for text sentiment and semantic analysis, statistical models and algorithms for classification and regression, network analysis, and so on. Meanwhile, only when solving these critical technical problems combined with domain theories can we discover knowledge in the domain of consumer behavior and decision support applications.

Therefore, we call for the research progress on consumer behavior and decision support applications in social commerce through data mining innovations in the field of mathematics and computer science.