Using children's search patterns to predict the quality of their creative problem solving

Purpose - The purpose of this paper is to propose a computational model that implicitly predict the children's creative quality of solutions by analyzing the query pattern on a problem-solving-based lesson. Design/methodology/approach - A search task related to the competencies acquired in the classroom was applied to automatically measure children' creativity. A blind review process of the creative quality was developed of 255 primary school students' solutions. Findings - While there are many creativity training programs that have proven effective, many of these programs require measuring creativity previously which involves time-consuming tasks conducted by experienced reviewers, i.e. far from primary school classroom dynamics. The authors have developed a model that predicts the creative quality of the given solution using the search queries pattern as input. This model has been used to predict the creative quality of 255 primary school students' solutions with 80 percent sensitivity. Research limitations/implications - Although the research was conducted with just one search task, participants come from two different countries. Therefore, the authors hope that this model provides detection of non-creative solutions to enable prompt intervention and improve the creative quality of solutions. Originality/value - This is the first implicit classification model of query pattern in order to predict the children' creative quality of solutions. This model is based on a conceptual relation between the concept association of creative thinking and query chain model of information search.