Vidwana

JOURNAL ARTICLE

Context-Aware Computer Science Professional Skills Recommender System: A Taxonomy-Based Chatbot Approach

Edward N. Udo1, Unwana I. Thomas1, Okure U. Obot1

1 Department of Computer Science, University of Uyo, Uyo, Akwa Ibom State, Nigeria

International Journal of Engineering and Computer Science, Vol. 15(06), pp. 28657-28665 · 2026-06-30 · DOI: 10.18535/ijecs/v15i06.5566

1

Reads

1

Downloads

0

Citations

Abstract

The development of a chatbot for context-aware professional skills recommendation in Computer Science (CS) addresses the need for personalized career guidance tailored to the dynamic nature of the field. The system employs advanced machine learning techniques, with Long Short-Term Memory (LSTM) models serving as the backbone for intent detection and context management. A structured taxonomy and knowledge base further enhance the system's ability to deliver precise recommendations, ensuring alignment with users' professional goals and skill needs. The implementation leverages the Flask Python web framework, chosen for its scalability, simplicity, and ability to support machine learning integration seamlessly. Key testing metrics showcase the chatbot's efficiency and accuracy, with a mean square error (MSE) of 2.042×10-6, an absolute error deviation that peaks at 0.002, and an average response time of 47 milliseconds. These results demonstrate a high degree of precision and responsiveness, making the system suitable for real-time use. The chatbot's design incorporates a user-friendly interface to facilitate accessibility and ease of use for both students and professionals, ensuring widespread applicability. Feedback from participants during the evaluation phase was largely positive, with users commending the system's ability to generate relevant and context-aware recommendations. This chatbot contributes significantly to the domain of professional development tools, offering a robust solution for bridging the gap between users’ aspirations and the skills required to achieve them. This work sets the foundation for further innovations in personalized learning technologies, particularly in areas requiring adaptive and context-aware capabilities.

Full text

Download PDF

Reader not loading? Open the PDF directly.

Cite this publication

Edward N. Udo, Unwana I. Thomas, Okure U. Obot (2026). Context-Aware Computer Science Professional Skills Recommender System: A Taxonomy-Based Chatbot Approach. International Journal of Engineering and Computer Science, 15(06), 28657-28665. https://doi.org/10.18535/ijecs/v15i06.5566
Download .ris (Zotero / Mendeley / EndNote)

Related publications