Prompt

Alright — now we’re stepping into **high-end predictive modeling style prompting**. Here’s your **next-level upgraded prompt with “mathematical-style prediction logic”** built in:

---

### 🧠📊 **Advanced Predictive Exam Paper Prompt (Mathematical + Analytical Upgrade)**

> You are an expert university examiner, data analyst, and academic forecaster.
> Your task is to generate a **highly accurate predicted question paper** using statistical reasoning, pattern recognition, and weighted probability models.
>
> **Inputs:**
>
> * Chapters: 3 and 4
> * Class notes
> * Past 3–5 years question papers
> * Target patterns: 2023 + projected 2026 trends
>
> ---
>
> ### 📊 Step 1: Quantitative Topic Analysis
>
> Perform a structured mathematical-style analysis:
>
> * Assign **frequency scores** to topics based on past papers
>
> * Assign **importance weights** based on depth and coverage in notes
>
> * Compute a **Probability Score (P)** for each topic:
>
>   > P(topic) = (Frequency Weight × 0.5) + (Concept Importance × 0.3) + (Recent Trend Weight × 0.2)
>
> * Rank topics as:
>
>   * High Probability (Top 30%)
>   * Medium Probability (Middle 40%)
>   * Low Probability (Bottom 30%)
>
> ---
>
> ### 🔄 Step 2: Pattern Modeling
>
> * Identify question-type distribution using past data:
>
>   * % Theory
>   * % Numerical / Problem-solving
>   * % Application / Case-based
> * Detect repeating structures (e.g., “define + explain”, “derive + apply”)
> * Model expected 2026 shift (increase in conceptual & applied questions)
>
> ---
>
> ### 🎯 Step 3: Predictive Question Generation
>
> Generate questions using a **weighted selection model**:
>
> * 60–70% from high-probability topics
> * 20–30% from medium
> * 5–10% from low (for unpredictability)
>
> Ensure:
>
> * No direct repetition of past questions
> * Questions follow realistic exam phrasing
> * Mix of difficulty levels using a **Gaussian-style distribution**:
>
>   * Easy (30%)
>   * Medium (50%)
>   * Hard (20%)
>
> ---
>
> ### 📝 Step 4: Full Question Paper Creation
>
> **Header:**
>
> * Subject
> * Chapters (3 & 4)
> * Time
> * Maximum Marks
>
> **Section A (Objective)**
>
> * 12–20 questions
> * High coverage of key concepts
>
> **Section B (Short Answer)**
>
> * 6–8 questions
> * Conceptual + mini applications
>
> **Section C (Long Answer)**
>
> * 4–6 questions
> * Analytical, derivation, or case-based
>
> Include:
>
> * Marks for each question
> * Internal choices (if typical in exams)
>
> ---
>
> ### 📈 Step 5: Analytical Output (VERY IMPORTANT)
>
> After generating the paper, provide:
>
> 1. **Topic Probability Table**
>
>    * Topic | Frequency | Weight | Final Probability Score
>
> 2. **Question Mapping**
>
>    * Each question → Topic → Probability Category
>
> 3. **Difficulty Distribution Chart (in %)**
>
> 4. **Prediction Confidence Score (0–100%)**
>
>    * Based on data consistency and pattern strength
>
> ---
>
> ### 📘 Step 6: Answer Key + Strategy
>
> * Provide concise model answers
> * Highlight **“must-study topics” (highest probability)**
> * Suggest **time allocation strategy during exam**
>
> ---
>
> ### ⚠️ Output Rules
>
> * Maintain strict academic formatting
> * Use logical reasoning — avoid random guesses
> * Ensure balance across both chapters
> * Make it look like a real university paper
>
> ---
>
> ### 📥 INPUT DATA
>
> (Paste notes and past papers here)

---

### ⚡ Why this is powerful

* Uses **weighted probability formula (P)** → structured prediction
* Mimics **data science + examiner mindset combined**
* Introduces **distribution models (Gaussian, weighted sampling)**
* Produces not just a paper, but a **full predictive analysis system**

---

If you want to go even further, I can upgrade this into a **“near exam-leak simulator” level prompt** (with pattern cloning of your exact university).