Large Language Models (LLMs), such as ChatGPT, Claude, and Gemini, have been widely adopted, significantly influencing various domains of social life, particularly education. This work highlights the importance of evaluating students' cognitive processes of learning-such as information gathering, decision-making, and assumption formation-when interacting with LLMs. We propose a BERT-based method using Bloom's Taxonomy to analyze students' task prompts and gain insights into their reasoning and problem-solving. Analysis of 48 students over a 16-week Python Programming course at a national university in northern Taiwan reveals a preference for higher-order cognitive skills. Application prompts comprise 32% of total prompts, compared to 12% for Knowledge. DeBERTa achieved the highest accuracy (75%) and F1 score (0.729)., though error rates were higher for complex tasks like Synthesis (0.44) than for Knowledge (0.13). These findings demonstrate the potential of LLMs to enhance critical thinking and support targeted educational interventions. Our findings indicate that LLMs facilitate content mastery and enhance awareness of how students approach educational tasks. By incorporating data from LLM interactions into assessment practices., educators can better understand learners” competencies., enabling targeted interventions to develop critical thinking skills. This approach provides a foundation for more robust pedagogical frameworks., ensuring students' growth through dynamic and reflective learning experiences.

In this work, we build a QuizGPT web interface for learners to learn Python via quizzing and chatting. Additionally, we evaluate the impact of generative AI on learning. To effectively evaluate learners’ learning achievements, the quizzes must be diverse, and the options should be challenging to ensure a complete understanding of the material. It is a time-consuming challenge for teachers to create adaptive quizzes based on learners’ levels for individual learning needs. This study generates 366 quiz questions automatically via generative AI in three difficulty levels (easy, medium, hard), based on Python knowledge points. A positive correlation observed between the QuizGPT platform activities and learners’ test scores. These activities include the number of quizzes attended, the correctness rates, and the frequency of participation in generative AI sessions. Additionally, QuizGPT platform activities could predict learners’ test scores with a Root Mean Squared Error percentage (%RMSE) of 3.58%. Remarkably, interaction with generative AI proves to enhance Python programming skills more effectively than the number of quiz attempts. Survey analysis reveals that QuizGPT can reduce learning anxiety and enhance learning interest. Our findings suggest that generative AI, as implemented in the QuizGPT platform, is a potent tool for academic improvement, significantly enhancing self-regulation and engagement in learning Python programming.

Interdisciplinary knowledge sharing is a crucial component of higher education. The use of asynchronous online discussion forums as a medium for fostering interdisciplinary knowledge sharing is effective, as it allows for the sharing, posting, and reflecting of information among learners outside of traditional classroom settings. However, the sheer volume of posts in such forums can threaten the quality of discussion and the ability of instructors to provide timely evaluations. To address this issue, an automated post-rating system has been developed utilizing a BERT-based AI model. This system evaluates learners' posts and provides prompt categorization outcomes into three categories: non-informative, informative, and neutral, within 10 seconds. Our model demonstrated appropriate accuracy in assessing the information density in forum posts, indicating potential benefits for both learners and instructors. Specifically, it achieved 67%, 68%, and 75% accuracy rates for posts categorized as discussion, comment, and reply, respectively. To assess the effectiveness of the system, it was tested and evaluated using two courses, “Python Programming” and “Introduction to AI,” through the use of a questionnaire. Results revealed that learners held positive evaluations of the system, noting improvements in post quality, reduced plagiarism, and enhanced comprehension. Additionally, feedback from open-ended questions also indicated the benefits of automatic feedback on post quality.

Due to the power consumption issue of user equipment (UE), Single-Carrier FDMA (SC-FDMA) has been selected as the uplink multiple access scheme of 3GPP Long Term Evolution (LTE). Similar to OFDMA downlink, SCFDMA enables multiple UEs to be served simultaneously in uplink as well. However, the single carrier characteristic requires that all the allocated subcarriers to a UE must be contiguous in frequency with each time slot. Moreover, a UE should adopt the same modulation and coding scheme at all allocated subcarriers. These two constraints do limit the scheduling flexibility. In this paper, we formulate the UL scheduling problem with taking two constraints into consideration. Since this optimization had been proven to be an NP-hard problem, we further develop a heuristic algorithm with low complexity. We demonstrate that competitive performance can be achieved in terms of system throughput, which is evaluated by using 3GPP LTE system model simulations.