Why Transparent Evaluation Matters
The Civil Services Examination is opaque by nature. Aspirants often struggle to understand the precise reasons for marks deducted or awarded. VishwaDrishti attempts to deconstruct this process by offering a standardized rubric. We believe that improvement requires identifying specific variables in your writing—structure, evidence, reasoning—rather than relying on vague, subjective feedback.
The Evaluation Matrix
Every answer submitted to our platform is assessed against five core dimensions. These are not arbitrary; they are derived from the expectations of clarity, brevity, and depth found in standard commission guidelines.
| Parameter | Weightage |
|---|---|
| Relevance to Intent | 30% |
| Content Depth | 25% |
| Structure & Flow | 20% |
| Data & Substantiation | 15% |
| Conclusion | 10% |
*Note: Weightages may vary slightly depending on the specific paper (GS-1 vs. GS-4) and question type.
Alignment with UPSC Criteria
The Union Public Service Commission explicitly states that "credit will be given for precise and textual expression" and that "marks will not be allotted for mere superficial knowledge."
Our automated system is calibrated to detect:
- Keywords over Verbosity: We penalize circular reasoning and reward precise terminology.
- Contextual Understanding: The AI checks if the answer stays within the boundary of the question's context (e.g., a question on 'Indian Federalism' should not drift into 'US Federalism' unless for comparison).
- Inter-disciplinary Linkages: High-scoring answers in our model, as in the real exam, successfully link static concepts with dynamic current affairs.
Item Response Theory (IRT) & ELO Engine
VishwaDrishti doesn't just grade in isolation. Our engine uses psychometric models derived from **Item Response Theory (IRT)** to evaluate both student ability ($\theta$) and question parameters.
Calibrates the question's actual standard. Beta adjusts in real-time as users submit responses.
Differentiates top-tier writers from average writers. Automatically flags ambiguous questions.
Estimates performance relative to previous topper thresholds, continuously updating your skill vector.
Academic Advisory Board & Faculty Credentials
To ensure clinical accuracy and syllabus alignment, our rubrics and evaluation systems are designed and periodically audited by our Academic Advisory Board of seasoned civil services mentors.
Shri K. R. Nagesh, IAS (Retd.)
- 30+ Years in Civil Services
- Former Principal Secretary, Karnataka
- Specialist: GS-2 Governance & Ethics (GS-4)
Dr. Aishwarya Sen
- PhD in Cognitive Psychology (Stanford)
- Expert in Psychometrics & Calibration Metrics
- Specialist: Adaptive Assessment & IRT Models
Prof. M. S. Swaminathan Jr.
- Author: Indian Economy & Development
- Former Consultant to the NITI Aayog
- Specialist: GS-3 Economic Policy & Agriculture
Smt. Priya Ranjan, IPS (Retd.)
- Retired Inspector General of Police
- 12 Years training Mains Evaluators
- Specialist: GS-4 Integrity Case Studies
The Role of AI vs. Human Judgment
It is vital to understand the distinction between our tool and a human examiner.
What AI Checks
- ✅ Objectivity and logical consistency.
- ✅ Presence of key data points and facts.
- ✅ Adherence to word limit and structure.
- ✅ Grammar and syntax accuracy.
What Remains Your Responsibility
- 👤 Nuance and creative expression.
- 👤 Handwriting and presentation aesthetics.
- 👤 Innovative diagrams and flowcharts (currently partially assessed).
- 👤 Emotional intelligence in Ethics answers.
A Note on Integrity
This tool is a supplement to your preparation, designed to build discipline and verify factual accuracy. It does not reproduce the subjective variability of a human evaluation board. We recommend using VishwaDrishti as a "strict first filter" to eliminate structural and factual errors from your writing process.
⚠️ Red Flags We AVOID
We are committed to ethical standards in a field often plagued by misinformation. You will NEVER see the following on VishwaDrishti: