Adaptive Learning Analytics and Personalized Curriculum Recommender System for Computer Education Students

Project Abstract

This study investigates an adaptive learning analytics framework integrated with a personalized curriculum recommender system designed for computer education students to enhance learning outcomes, engagement, and self-regulated learning. By leveraging multimodal data sources including learner interactions, assessment results, time-on-task, and content engagement signals, the framework builds student profiles that capture cognitive, affective, and metacognitive dimensions. A hybrid analytics pipeline combines descriptive, predictive, and prescriptive analytics to identify learning gaps, predict performance trajectories, and generate individualized curriculum pathways aligned with national standards and programmatic objectives. The recommender component utilizes collaborative filtering, content-based similarity, and knowledge tracing augmented with Bayesian networks to suggest sequenced modules, hands-on activities, and formative assessments tailored to each learner’s current mastery level, preferred learning style, and career aspirations. The system supports dynamic adaptation at multiple granularity levels, including course-wide progression, topic-level mastery, and micro-competency targets, while ensuring pedagogical coherence and alignment with instructor-crafted syllabi. A novel weighting scheme balances accuracy, explainability, and fairness, enabling transparent recommendations that instructors can monitor and customize. The research employs a mixed-methods approach, combining quantitative experiments with qualitative feedback from students and educators across multiple computer education courses. Quantitative evaluation focuses on learning gains measured by validated performance rubrics, time-to-competence, and retention rates, as well as system accuracy in identifying at-risk students. A quasi-experimental design compares cohorts using the adaptive system against control groups following traditional curricula, supplemented by A/B testing of recommendation strategies. Qualitative analysis explores perceived autonomy, motivation, perceived usefulness, and spo?ecz-emotional factors associated with personalized learning pathways. The study also examines the system’s impact on curriculum coverage, resource utilization, and instructor workload, with attention to scalability and interoperability within existing learning management systems. Data privacy and ethical considerations are addressed through differential privacy mechanisms, secure data governance, and user consent protocols. Results are expected to demonstrate statistically significant improvements in learner achievement, higher engagement metrics, and more efficient progression through competencies, alongside enhanced satisfaction with the learning experience. The research contributes to theories of adaptive learning and micro-credentialing by integrating robust analytics with actionable curriculum recommendations, offering a replicable architectural blueprint for institutions seeking to personalize computer education at scale. Limitations include potential biases in data collection, the need for domain-specific calibration of knowledge traces, and challenges in capturing deep conceptual understanding through proxy indicators. Recommendations for future work include extending the framework to multidisciplinary contexts, integrating advanced natural language processing for feedback, and exploring real-time intervention strategies to sustain long-term learning momentum.

Project Overview

What This Project Is About

---

A straightforward exploration of how learning analytics can tailor a computer education course to each student’s needs. The project looks at gathering information about how students study, what concepts they struggle with, and how to suggest personalized activities and readings to improve understanding and engagement.

---

---

The Problem It Addresses

---

Many computer education courses use one-size-fits-all teaching materials. This can leave students behind or bored. The project aims to bridge this gap by using simple data about study patterns to adjust the learning path for individual students.

---

---

Objectives of the Project

---

1. Understand what data about learning is useful and ethical to collect.
2. Develop a basic system to track student progress and identify knowledge gaps.
3. Create a simple recommendation method that suggests next activities and readings.
4. Evaluate whether personalized recommendations improve understanding and motivation.
5. Provide guidelines for educators on using the tool in class.

---

---

What You Will Do Step by Step

---

1) Review simple learning analytics ideas and select practical metrics. 2) Design a lightweight data collection plan (quizzes, time-on-task, and activity types). 3) Build a minimal prototype that suggests activities based on collected data. 4) Test with a small group of students and collect feedback. 5) Analyze results to see if the personalized path helps learning. 6) Document the process and results with clear instructions for reuse.

---

---

---

---

Expected Outcome

Expect a functioning prototype that offers personalized activity recommendations and a report on whether these suggestions support better learning outcomes in computer education courses. The project should produce user-friendly guidelines for teachers and a path for future improvements.