GoldMind: A Teacher-Centered Knowledge Management System for Higher Education -- Lessons from Iterative Design

TL;DR

The paper tackles the challenge of effectively capturing and transferring teaching knowledge in higher education amid high staff turnover. It presents GoldMind, a teacher-centered KMS developed and evaluated through a three-iteration, human-centered design study with 108 university teachers, leveraging process mining and Generative AI for in-the-flow knowledge capture and dissemination. Key contributions include empirical insights into technological, design, and human-factor factors shaping KMS adoption, the ShareFlows visualization for expert-to-novice transfer, and ENA-based analysis of knowledge-management behaviours. The findings demonstrate improved retrieval speed, knowledge capture efficiency, task completion, and usability, illustrating the value of iterative, participatory design for usable and trusted AI-enabled KMS in education and outlining directions for broader, longitudinal validation.

Abstract

Designing Knowledge Management Systems (KMSs) for higher education requires addressing complex human-technology interactions, especially where staff turnover and changing roles create ongoing challenges for reusing knowledge. While advances in process mining and Generative AI enable new ways of designing features to support knowledge management, existing KMSs often overlook the realities of educators' workflows, leading to low adoption and limited impact. This paper presents findings from a two-year human-centred design study with 108 higher education teachers, focused on the iterative co-design and evaluation of GoldMind, a KMS supporting in-the-flow knowledge management during digital teaching tasks. Through three design-evaluation cycles, we examined how teachers interacted with the system and how their feedback informed successive refinements. Insights are synthesised across three themes: (1) Technology Lessons from user interaction data, (2) Design Considerations shaped by co-design and usability testing, and (3) Human Factors, including cognitive load and knowledge behaviours, analysed using Epistemic Network Analysis.

Figures (7)

• Figure 1: Human-centred iterative design and implementation of the GoldMind Knowledge Management System. Each iteration involved targeted activities, beginning with benchmarking existing KMSs (M6), followed by co-design sessions to understand teachers' workflows and tasks (M7), user evaluations (M12), and qualitative assessments in near-authentic contexts (M18). The final implementation and large-scale evaluation occurred at M24. Across these phases, features evolved incrementally, including Knowledge Repository, Knowledge Capture in the Flow (v1 to v2), Contextual Highlights, ShareFlow visual representations, and automated recommendation features.
• Figure 2: Representative teaching tasks performed by higher education teachers in their teaching workflow. Red indicates those tasks that are performed more than once per semester. We underscore the tasks selected for evaluation in this study. Task complexity is measured by the time and the number of resources required to complete the task.
• Figure 3: GoldMind Features. The diagram illustrates key components that support knowledge capture and dissemination within the GoldMind platform. Features include: (A) a centralised Knowledge Repository; (B) in-the-flow Knowledge Capture, including knowledge clipping, annotations, and uploading personal knowledge via a browser extension; (C) video annotation capture; (D) Visual Representation of Process Flows through ShareFlows; (E) contextual recommendations powered by highlights and summaries; (F) automated ShareFlow push recommendations; and (G) automated contextual recommendation mechanisms that surface relevant knowledge at the point of need.
• Figure 4: System Usability survey response rating by participants in Iteration 1
• Figure 5: The results of Iteration 1 (M6) -- the descriptive statistics of the responses to the NASA task load index survey