Problems in digital public services — Technical https://thesis.uya.no/proposals/spot-the-digital-roadblocks/#technical-track BACHELOR Final result: A reproducible triage prototype and error report. Task: Build a TF–IDF logistic comment classifier with confidence, abstention and a manual correction log. Data: 200 fictional comments + 40 pilot comments (planned) Five service steps; 180 single-topic and 20 mixed/unclear evaluation comments. Label step and actionable problem separately. Vary vocabulary, tone and length independently of labels. Requires: Independent label checking and IS students; all comments must be authored. Method: Use the reviewed synthetic comments, paraphrase-family splits and keyword/majority baselines; report held-out errors and correction effort in 4 novice-analyst sessions. STEPS Build and test 1. Read the starting sources and choose one established implementation method. 2. Write the requirements, data fields, system diagram and test cases. 3. Prepare 200 fictional comments + 40 pilot comments. Make the answer key and pass criteria before testing. 4. Build a working version of a reproducible triage prototype and error report. 5. Run function, integration and failure-case tests. Record each result. 6. Run the practical evaluation and list the changes the system still needs. MASTER Final result: Versioned taxonomy, supervised triage service, review dashboard and held-out evaluation. Research question: Investigate how ambiguity changes human-in-the-loop feedback triage. Compare assisted and manual triage on single- and mixed-problem comments; test whether correction recovers errors without introducing new ones. Research result: Design principles for calibrated, reviewable public-service feedback triage under ambiguity. STEPS Design, build and test (DSR) 1. Read the newest papers and list the closest existing systems. 2. Write down the versions, fields, data split, case assignment, random seeds and correct answers for 200 fictional comments + 40 pilot comments. 3. Draw the user workflow, data model and system architecture. List the requirements and pass criteria. 4. Build a working version of versioned taxonomy, supervised triage service, review dashboard and held-out evaluation. 5. Test every function, connection and failure case. Save the failed tests as well as the passed tests. 6. Compare the system with the named alternative. Then run the user task or decision task in the assignment. 7. Report the measured result, the failed cases and the design lessons another team can reuse. CURRENT PROJECT LITERATURE Prioritising national healthcare service issues from free text feedback (2024, peer-reviewed): https://doi.org/10.1016/j.dss.2024.114215 Shows how text analytics can support explainable prioritization of service-improvement decisions. Designing a digital citizen-centered service (2024, peer-reviewed): https://doi.org/10.1016/j.techfore.2024.123280 Positions citizen feedback and stakeholder involvement inside a digital-service design process. IS THEORY STARTING POINTS Amershi et al. (2019) — Guidelines for Human-AI Interaction: https://doi.org/10.1145/3290605.3300233 Select and justify the correction, explanation or uncertainty design being tested. Wang & Strong (1996) — Beyond Accuracy: https://www.tandfonline.com/doi/abs/10.1080/07421222.1996.11518099 Distinguish accuracy, completeness, representation and fitness for the task. Search Scopus or Web of Science and ACM Digital Library using the topic query, then follow citations to the thesis start date. Record searches and compare methods, data, findings and limitations in literature-matrix.csv. Use that review to confirm or revise the gap and choose a current comparator. The linked papers are starting points. PILOT QUESTION Does evidence-linked feedback grouping help service owners identify specific problems to fix? EXAMPLE TOOLS Python, pandas, scikit-learn, Streamlit, pytest Equivalent tools are fine. FIRST THREE ACTIONS 1. Write 10 comments across the five service steps; label the step and zero or more problem types separately. 2. Define the multilabel taxonomy and keep every paraphrase family in one train, validation or test partition. 3. Run the majority and keyword-rule baselines first, then train the two fixed logistic-regression pipelines without inspecting test labels. LITERATURE SEARCH digital public service feedback analysis usability barriers DATA 200 fictional comments + 40 pilot comments (planned) Five service steps; 180 single-topic and 20 mixed/unclear evaluation comments. Group paraphrases into 120/40/40 train/validation/test partitions. Label one step plus zero or more problem types: navigation, information clarity, validation/error, availability, confirmation, accessibility and other/unclear. HOW TO TEST IT Tune logistic C and multilabel thresholds on validation only, then finalise. Report macro/micro-F1, exact match, mixed-comment recall, Brier/ECE and per-label confusion on test. Six novice IS analysts triage 20 non-overlapping comments in assisted versus manual views; measure correction recovery, introduced errors, missed actionable issues and time. FINAL RESULT Versioned taxonomy, supervised triage service, review dashboard and held-out evaluation. ACCESS OR PEOPLE Independent label checking and 6 IS students; all comments must be authored. TEMPLATES https://thesis.uya.no/starters/spot-the-digital-roadblocks/technical/data-plan.csv https://thesis.uya.no/starters/spot-the-digital-roadblocks/technical/evaluation.csv https://thesis.uya.no/starters/spot-the-digital-roadblocks/technical/literature-matrix.csv Use participant codes instead of names or email addresses.