This project utilizes data from the abstracts of 82 articles included in a systematic review of cybersecurity education for K-12 students.
To prepare the data for tokenization, the text will be cleaned by converting all words to lowercase and removing stems, ensuring greater accuracy in identifying high-frequency words.
The bar plot illustrates an overall upward trend in cybersecurity publications, despite some fluctuations in the middle years. Notably, 2024 shows a significant increase in publications, marking a sharp rise compared to previous years.
The word cloud captures the most frequnelty otryccuring terms across all articles. These words tend to represent general terminology in cybersecurity education, such as online safety, digital citizenship, cybersecurity awareness, digital literacy, and parent mediation. However, it does not reflect distinctive themes in this set of articles.
Based on the TF-IDF results, many article-specific unique words or phrases have been identified, such as comic book, preservice teacher, objective knowledge, protective skill, online communication, sexual exploitation, online market, and cartoon video. These findings contrast significantly with the previous high-frequency wordcloud list, which primarily highlighted commonly discussed topics in the field. When combined with my human coding results, it becomes evident that many of these terms align with specific themes, such as innovative instructional strategies or learning outcomes of cybersecurity education.