Bonaboss01/Trump-Nigeria-Sentiment-Analysis

NLP project analyzing global reactions to Donald Trump’s 2025 statement on Nigeria using sentiment analysis, topic modeling, and named entity recognition. NLP project analyzing global reactions to Donald Trump’s 2025 statement on Nigeria using sentiment analysis, topic modeling, and named entity recognition. # 🧠 Trump Nigeria Sentiment Analysis ** Natural Language Processing (NLP) project analyzing global reactions to Donald Trump’s 2025 statement on Nigeria using sentiment analysis, topic modeling, and named entity recognition.** --- ## 📊 Data Collection I initially planned to scrape tweets related to Donald Trump’s 2025 statement on Nigeria using the `snscrape` Python library. However, due to recent compatibility issues with Python 3.13, `snscrape` produced import errors that prevented execution. I explored several alternative approaches, including: - **Tweepy (Twitter API v2):** Provides live tweet collection but requires a valid API key and is rate-limited for free accounts. - **NewsAPI / GDELT:** Could collect related news articles but would shift the project’s focus away from individual opinions to media coverage. - **Pre-existing open-source Twitter datasets:** Contained unrelated or outdated topics. Given that the purpose of this project was to **apply and demonstrate NLP techniques** — not to test API integrations — I opted to generate a **synthetic dataset** that realistically mirrors social media discourse on the topic. The dataset includes **2,000 samples** across multiple perspectives: - pro-Trump - anti-Trump - neutral - fact-checking - sarcastic - sovereignty-defending tweets Each record includes realistic metadata such as **timestamps**, **likes**, and **retweets**. --- ### ✅ Advantages of This Approach - **Reproducibility:** Anyone can run the notebooks without API credentials. - **Control:** Balanced sentiment classes and representativeness across perspectives. - **Continuity:** Enabled smooth progression to the analytical workflow — text cleaning, sentiment analysis, topic modeling, and named entity recognition. ---