Abdikani94/dareen-somali-sentiment-ai

Somali-first AI sentiment analyzer with negation, spelling-variation, and contextual language support. # Dareen **Somali-first sentiment analysis for Somali and English feedback.** Dareen classifies text as positive, neutral, or negative while accounting for Somali negation, spelling variation, word forms, and sentence context. ## Features - Somali and English language detection - Positive, neutral, and negative classification - Confidence score and topic detection - Negation-aware Somali analysis (`ma`, `aan`, `kama`, and related forms) - Spelling normalization and morphological matching - Evidence phrases explaining each result - Responsive, accessible interface - Optional OpenAI analysis with a local fallback engine ## Analyzer ## Results | Positive | Neutral | Negative | | --- | --- | --- | | | | | ## Run locally Requirements: Node.js 22.13 or newer. ```bash npm install npm exec vinext dev ``` Open localhost. To enable OpenAI-powered analysis, create `.env.local`: ```env OPENAI_API_KEY=your_key_here ``` The app works without an API key by using its local context engine. ## Tech stack Next.js · TypeScript · React · Vinext · Tailwind CSS · OpenAI API ## Author Built by Abdikani Mohamed.