{"id":50514171,"url":"https://github.com/krish902/question-paper-difficulty-analyzer","last_synced_at":"2026-06-02T22:31:46.267Z","repository":{"id":356648423,"uuid":"1233484324","full_name":"krish902/question-paper-difficulty-analyzer","owner":"krish902","description":"Automated question paper quality analysis using BERT and Bloom's Taxonomy - published in 2 Springer conferences","archived":false,"fork":false,"pushed_at":"2026-05-09T02:59:12.000Z","size":74,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-09T04:36:58.727Z","etag":null,"topics":["bert","blooms-taxonomy","nlp","python","tensorflow"],"latest_commit_sha":null,"homepage":"https://link.springer.com/chapter/10.1007/978-981-97-8526-1_6","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/krish902.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-05-09T02:32:53.000Z","updated_at":"2026-05-09T03:19:33.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/krish902/question-paper-difficulty-analyzer","commit_stats":null,"previous_names":["krish902/question-paper-difficulty-analyzer"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/krish902/question-paper-difficulty-analyzer","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/krish902%2Fquestion-paper-difficulty-analyzer","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/krish902%2Fquestion-paper-difficulty-analyzer/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/krish902%2Fquestion-paper-difficulty-analyzer/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/krish902%2Fquestion-paper-difficulty-analyzer/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/krish902","download_url":"https://codeload.github.com/krish902/question-paper-difficulty-analyzer/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/krish902%2Fquestion-paper-difficulty-analyzer/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33840213,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-02T02:00:07.132Z","response_time":109,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bert","blooms-taxonomy","nlp","python","tensorflow"],"created_at":"2026-06-02T22:31:45.056Z","updated_at":"2026-06-02T22:31:46.249Z","avatar_url":"https://github.com/krish902.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Bloom's Taxonomy Question Difficulty Classifier using BERT\n\nAutomated NLP system that classifies university exam questions into Bloom's Taxonomy cognitive levels and predicts overall question paper difficulty (Easy / Medium / High).\n\nPublished across **2 Springer conference proceedings** — ICT4SD 2024 and AITA 2024.\n\n---\n\n## Results\n\n| Model | Validation Accuracy |\n|-------|-------------------|\n| LSTM | 48.76% |\n| Bi-LSTM | 67.32% |\n| **BERT (ours)** | **98.52%** |\n\nBERT significantly outperforms traditional deep learning approaches by capturing long-range semantic dependencies in question text.\n\n---\n\n## What It Does\n\n1. Takes a university exam question as input\n2. Classifies it into one of 6 Bloom's Taxonomy levels:\n   - `0` Remember → `1` Understand → `2` Apply → `3` Analyze → `4` Evaluate → `5` Create\n3. Aggregates question-level scores using a weighted formula to predict overall paper difficulty\n\n---\n\n## Dataset\n\n- **3,510 questions** collected from university exam papers across multiple subjects\n- Two columns: `Questions` and `Bloom's Taxonomy` label\n- Labels encoded as: Remember=0, Understand=1, Apply=2, Analyze=3, Evaluate=4, Create=5\n\nSample:\n\n| Question | Bloom's Level |\n|----------|--------------|\n| How many steps are required to solve Tower of Hanoi? | Remember |\n| Design an algorithm to detect cycles in a graph. | Analyze |\n| What is Cyclomatic complexity? | Understand |\n\n---\n\n## Tech Stack\n\n- **Model:** BERT (`bert-base-cased`) via HuggingFace Transformers\n- **Framework:** TensorFlow / Keras\n- **Tokenization:** BertTokenizer, max length 256\n- **Training:** Adam optimizer, ReLU activation, 30 epochs, batch size 16\n- **Evaluation:** Accuracy, Precision, Recall, Confusion Matrix\n\n---\n\n## Project Structure\n\n```\nquestion-paper-difficulty-analyzer/\n├── blooms_level_final_with_matrices_inc_data.ipynb  # Full training pipeline\n├── questions.csv                                    # Dataset of 3,510 labelled questions\n├── requirements.txt\n└── README.md\n```\n\n---\n\n## How to Run\n\n```bash\ngit clone https://github.com/krish902/question-paper-difficulty-analyzer\ncd question-paper-difficulty-analyzer\npip install -r requirements.txt\njupyter notebook blooms_level_final_with_matrices_inc_data.ipynb\n```\n\n---\n\n## Publications\n\nThis work is published in two peer-reviewed Springer conference proceedings:\n\n1. **Impact of BERT on Evaluating the Quality of Question Papers using Bloom's Taxonomy**\n   ICT4SD 2024 — 9th International Conference on ICT for Sustainable Development, Goa, India\n\n2. **Difficulty Level Prediction on Evaluating the Quality of Question Papers using Bloom's Taxonomy**\n   AITA 2024 — 2nd International Conference on Artificial Intelligence: Theory and Applications, Bangalore, India\n\n---\n\n## Authors\n\nKrish Bhikadiya, Dhaval Patel, Hemit Rana, Nikita Bhatt\nChandubhai S. Patel Institute of Technology, CHARUSAT, India\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkrish902%2Fquestion-paper-difficulty-analyzer","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkrish902%2Fquestion-paper-difficulty-analyzer","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkrish902%2Fquestion-paper-difficulty-analyzer/lists"}