MSc Financial NLP Module
Corporate Financial Reports & SEC Filings NLP
Automated semantic parsing, chunk-level FinBERT classification, and section-wise risk analysis for Form 10-K, 10-Q & Earnings Transcripts.
Hierarchical NLP Document Ingestion Pipeline
Long-form statutory filings exceed standard transformer context windows (512 tokens). ArthVeda segments statutory documents into thematic disclosures (Item 1A Risk Factors, Item 7 MD&A, and Earnings Transcripts), runs sliding-window contextual classification via ProsusAI/finbert, and synthesizes a domain-weighted corporate health index.
Parsing statutory SEC filings & computing FinBERT cross-attention embeddings...