Supervised diagnostic classification of cognitive attributes using data augmentation.

Over recent decades, machine learning, an integral subfield of artificial intelligence, has revolutionized diverse sectors, enabling data-driven decisions with minimal human intervention. In particular, the field of educational assessment emerges as a promising area for machine learning applications, where students can be classified and diagnosed using their performance data. The objectives of Diagnostic Classification Models (DCMs), which provide a suite of methods for diagnosing students' cognitive states in relation to the mastery of necessary cognitive attributes for solving problems in a test, can be effectively addressed through machine learning techniques. However, the challenge lies in the latent nature of cognitive status, which makes it difficult to obtain labels for the training dataset. Consequently, the application of machine learning methods to DCMs often assumes smaller training sets with labels derived either from theoretical considerations or human experts. In this study, the authors propose a supervised diagnostic classification model with data augmentation (SDCM-DA). This method is designed to utilize the augmented data using a data generation model constructed by leveraging the probability of correct responses for each attribute mastery pattern derived from the expert-labeled dataset. To explore the benefits of data augmentation, a simulation study is carried out, contrasting it with classification methods that rely solely on the expert-labeled dataset for training. The findings reveal that utilizing data augmentation with the estimated probabilities of correct responses substantially enhances classification accuracy. This holds true even when the augmentation originates from a small labeled sample with occasional labeling errors, and when the tests contain lower-quality items that may inaccurately measure students' true cognitive status. Moreover, the study demonstrates that leveraging augmented data for learning can enable the successful classification of students, thereby eliminating the necessity for specifying an underlying response model.

The Cyclic Value-Context Reinforcement Model of Problematic Internet Use: Empirical Validation Using a Thematic Analysis of Children's Counseling Data.

BACKGROUND: Research on problematic internet use has focused on devising diagnostic criteria or describing the factors that influence internet overuse. However, a paradigm shift is necessary in studying the phenomenon of increased internet use not just from a pathological point of view but also from a developmental point of view that considers children's behavior of adapting to a technology-oriented society. OBJECTIVE: In this paper, we propose the Cyclic Value-Context Reinforcement Model (CVCRM) to understand problematic internet use behavior. The purpose of our study was to construct a developmental process model that provides a holistic understanding of problematic internet use behavior of children and to empirically validate the proposed model by conducting a thematic analysis on actual counseling data. METHODS: To validate the CVCRM, we conducted thematic analysis using the counseling data from 312 Korean children aged 7-18 years. For the coding process, 7 master's and doctoral student researchers participated as coders, and 2 professors supervised the coding process and results. RESULTS: This project was funded from October 2015 to September 2019 to analyze counseling data from 312 children who participated in counseling sessions during January 2012 to May 2014. Based on the data analysis, we present the CVCRM, which integrates existing theoretical approaches and encompasses the 3 interacting aspects that induce and reinforce problematic internet use in children: psychosocial value, environmental context, and internet utility. Specifically, using counseling data, we empirically ascertained that problematic internet use behavior feeds into children's psychosocial values and environmental contexts, which in turn facilitates problematic internet use in a cyclical manner. CONCLUSIONS: Through this empirical validation, the CVCRM can provide a theoretical framework and an integrated perspective on the developmental mechanism of problematic internet use behavior of children.

Classification of problematic Internet usage types by motives and contexts with elementary and secondary school-aged counseling clients.

BACKGROUND AND AIMS: Past research on the classification of problematic Internet use (PIU) has focused on symptom-based severity and usage motive in order to understand its mechanism. Recently, usage context, such as family or social relationships, has been identified as a key influencing factor of PIU. Therefore, we extended the classification of PIU to include usage context in addition to symptom-based severity and usage motive. METHODS: To classify PIU types, we conducted two studies. First, we performed a clustering analysis, where 265 counseling cases were clustered into usage types in terms of motive and context. After characterizing each usage type, we examined their hierarchical relationships by considering symptoms. Second, we performed a focus group interview with six counselors to increase the transferability of usage types. This transferability was established by matching counselors' quotations to the usage types. When usage types showed consistency between quotations and the clustering analysis, we identified the progression patterns between hierarchical relationships. RESULTS: The clustering analysis of motive and context yielded six usage types with three hierarchical relationships. The focus group interview results verified the transferability of these six types and identified two progression patterns between the hierarchical relationships. DISCUSSION AND CONCLUSION: Although usage motive and context were given equal weight before the clustering analysis, the resulting types revealed that usage context played a greater role in the classification process.