Smart Study Advisor

A course recommendation system built with multiple programming paradigms. Offers two recommendation pipelines: a non-AI path powered by a Prolog inference engine, and an AI path using Google Gemini.

Paradigm Usage

Paradigm	Where	Why
Logic (Prolog)	api/prolog/rules.pl	defines course facts, prerequisite chains, and recommendation rules.
OOP	api/models.py	Django models (Major, Course) encapsulate data with relationships.
Functional	api/ai_is_calling.py	Data transformation pipeline using map/lambda for normalizing topics, pure functions (build_prompt, parse) that take input and return output with no side effects.
Imperative	api/prolog_engine.py, api/views.py	Sequential execution flow assert facts, run queries in order, collect results, clean up.

Prolog Knowledge Base

api/prolog/rules.pl contains:

• Course facts: course(Name, Difficulty, Topic, Major) ~300 courses across 20+ majors
• Prerequisite facts: prerequisite(Course, PrerequisiteCourse)
• Dynamic predicates: completed/1, student_preference/1, prefers_difficulty/1, student_major/1 asserted per-request by the Python engine
• Helper rules:
  • major_matches/1 if no major is asserted, matches any major
  • effectively_completed/1 if you completed math2, math1 is inferred as completed since it's a prerequisite

Recommendation Fallback Chain

The engine tries progressively relaxed queries:

1. Strict: matches topic + difficulty + major
2. Drop topic: keeps difficulty + major
3. Drop difficulty: keeps topic + major
4. Drop both: major only

Each level has two clauses: one for courses with prerequisites (checks all are met), one for courses with no prerequisites.

API Endpoints

Method	Endpoint	Description
GET	/api/majors/	List all majors
GET	/api/courses/?major=<name>	List course names for a major
POST	/api/recommend/prolog/	Prolog recommendation
POST	/api/recommend/ai/	AI recommendation (Gemini)

POST /api/recommend/prolog/

{
    "liked_topics": ["mathematics", "algorithms"],
    "completed_courses": ["math1", "math2"],
    "preferred_difficulty": "hard",
    "major": "computer_and_systems"
}

POST /api/recommend/ai/

Same request body. Uses Gemini to generate recommendations based on a structured prompt that includes the prerequisite graph from the Django database.

Setup

Backend

cd backend
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Requires SWI-Prolog installed on the system (sudo apt install swi-prolog).

Create backend/.env:

GEMINI_API_KEY=your_key_here

python manage.py migrate
python manage.py runserver

Mobile App

npm install
npx expo start

Tech Stack

• Backend: Django + Django REST Framework
• Prolog Engine: SWI-Prolog via PySwip
• AI: Google Gemini API (gemini-2.5-flash)
• Mobile: Expo / React Native (TypeScript)
• Database: SQLite