Python / Machine Learning Developer Needed – Fake News Detection FYP

*Python / Machine Learning Developer Needed – Fake News Detection System FYP* Project Title: Fake News Detection System I am looking for an experienced Python and Machine Learning developer to help me develop my complete Final Year Project. I have an approved project proposal and an official requirements document from my supervisor. The implementation must follow those requirements. *Project Overview* Develop a Python-based Fake News Detection System using the Kaggle Fake and Real News Dataset (`Fake.csv` and `True.csv`). The system should classify news articles as **Potentially Fake** or **Potentially Genuine** using patterns learned from the dataset. It must not claim to independently verify whether an article is true or false. *Required technologies:* Python, Pandas, NumPy, NLTK, Scikit-learn, Matplotlib, Seaborn, and Flask or Streamlit. *Required Functionalities* All 10 functional requirements must be implemented: 1. Dataset loading and exploration. 2. Data preprocessing and cleaning. 3. TF-IDF feature extraction. 4. Proper training/testing data splitting. 5. Training and comparison of Logistic Regression, Multinomial Naive Bayes, and SVM. 6. News article classification. 7. Evaluation using Accuracy, Precision, Recall, F1-score, and Confusion Matrix. 8. Robustness testing and error analysis. 9. Graphs and data visualizations. 10. User-friendly interface and error handling. *Important:* Use TF-IDF and the three specified ML algorithms. Do not replace them with BERT or other deep-learning models. *Deliverables and Payment Structure* *Four milestones – escrow protected* *Milestone 1: $25 – Foundation* - Software Requirements Specification (SRS). - Software Design Document (SDD). - UML diagrams: Use Case, Activity, Sequence, and Class. - Data Flow Diagrams (DFD). - Documentation must accurately reflect the actual implementation and be updated if the design changes. *Milestone 2: $30 – Core Machine Learning* - Dataset loading, exploration and preprocessing. - TF-IDF feature extraction. - Training and comparison of Logistic Regression, Multinomial Naive Bayes and SVM. - Model evaluation, testing, error analysis and graphs. *Milestone 3: $25 – Complete Application* - Initial working prototype. - Complete Flask or Streamlit application. - Potentially Fake / Potentially Genuine prediction output. - Appropriate disclaimer. - Input validation and error handling. - `requirements.txt` file. - Installation and setup guide. - Application screenshots. *Milestone 4: $20 – Final Handover and Viva Preparation* - Final project report. - Complete source code and editable documentation. - Full code walkthrough in simple English. - At least 100 potential viva questions with answers. - Help with installation, running, and demonstrating the project. *Important Conditions* - Each milestone must be delivered for my review and approval before payment is released. - Payments must be handled through the platform's escrow/milestone protection. - All source code and documentation must be original and properly referenced. - I must receive all project files and be able to run the application independently. - I am a beginner in programming, so explanations must be provided in simple English. - I need to understand the implementation well enough to explain and demonstrate it during my viva. - Work must comply with my university's academic-integrity requirements. *Please Include in Your Proposal* 1. Your experience with Python, NLP, and Machine Learning. 2. Examples of similar projects. 3. Confirmation that you can meet all 10 functional requirements. 4. Confirmation that documentation and viva preparation are included. 5. Estimated completion time. 6. Confirmation that you accept the $100 budget and four-milestone payment structure. To prove you read the full spec, start your proposal with "FYP-FNDS-ORIGINAL - TF-IDF Only, No BERT". Proposals without this will be declined.