Overview

Luna Web Platform Overview

Mission & Scope

Luna is a research-driven web platform that partners with universities to understand and mitigate mathematics student dropout. It combines structured wellbeing data collection, real-time analytics, and actionable insights so that researchers, lecturers, and support staff can coordinate interventions before students disengage.

This document provides a high-level orientation to the platform for new collaborators, grant partners, and technical contributors.

Stakeholders

Role	Primary Goals	Luna Support
Researchers	Capture longitudinal wellbeing data and evaluate retention hypotheses	Automated survey pipeline, raw/processed data exports, reproducible modelling jobs
Lecturers	Monitor module health and respond to concerning trends	Cohort dashboards, risk alerts, module configuration tools
Students	Receive timely check-ins and track personal wellbeing trends	Weekly micro-surveys, personal analytics, privacy controls
Platform Engineers	Operate the infrastructure reliably and securely	Containerised services, observability hooks, automated cron scheduling

Value Proposition

• Evidence-based retention insights – captures both background and weekly longitudinal signals, then applies Kalman-filter smoothing to detect at-risk trajectories early.
• University-ready operations – supports multi-university, multi-module deployments with lecturer-controlled enrolment and compliance-friendly data segregation.
• Explainable analytics – exposes cohort-level trends and per-student narratives that are rooted in transparent psychometric models rather than black-box scoring.
• Extensible architecture – API-first Django backend, modular modelling package, and documented analytics outputs simplify further research integrations.

Core Capability Areas

1. Identity & Access Management – custom user model with student, lecturer, and administrator roles, university scoping, and Django admin support.
2. Module & Survey Lifecycle – background forms, password-protected module enrolments, scheduled weekly surveys, and reminder cron jobs.
3. Analytics Pipeline – modelling app orchestrates Kalman filtering and risk scoring, persisting smoothed metrics for dashboards and exports.
4. Notifications & Reporting – automated cron runners execute survey creation and analytics updates; optional email hooks are available for interventions.
5. Operations & Observability – containerised deployment (Docker + docker-compose), PostgreSQL persistence, and PGAdmin for manual inspection.

Platform Components

Component	Description	Key Technologies
Backend API	RESTful endpoints for account management, module orchestration, forms, surveys, and analytics	Django 4.2.x, Django REST Framework
Modelling Service	Processes survey streams, performs Kalman filtering, and stores smoothed trajectories	Python data stack, custom modelling package
Task Scheduling	Ensures surveys are generated, processed, and escalated on time	django-cron, custom cron runner loop
Data Layer	Source of truth for all user, module, form, and survey data	PostgreSQL 14, Django ORM
Ops Tooling	Local and production observability, DB management, deployment automation	Docker, Docker Compose, PGAdmin, Makefile helpers

Feedback Channels

• GitHub Issues – for bugs and feature requests (tag with web-platform).
• Research Steering Group – meets monthly to review analytics findings and prioritise new instrumentation.
• Security & Compliance – report incidents to the platform administrators; SOC-style runbooks are documented in the private operations repository.

📚 Continue with the companion guides:

• web-platform/installation.md for developer setup instructions.
• web-platform/architecture.md for infrastructure, data modelling, and request flows.
• web-platform/student-experience.md and web-platform/lecturer-experience.md for persona-centred walkthroughs.