DealForge autonomously sources, scores, and writes investment memos on venture deals. Stop manually hunting.
1,180+ deals tracked · 22 AI investment memos · Updated daily
Show HN: Vocab extractor for language learners using Stanza and frequency ranks
I'm building a Telegram bot to practice Dutch. GPT-4o-mini kept picking vocabulary words I already knew, so I built a classical NLP pipeline to do it instead.<p>It takes a short text + learner level (A0–B1) and returns the best words to study, using Stanza for parsing and corpus frequency ranks (SUBTLEX-NL, srLex, SUBTLEX-US) for scoring. Wins at A1/A2, loses at A0 where the LLM picks more obvious words.<p>I also tried adding multi-word phrases (ADJ+NOUN, VERB+NOUN, phrasal verbs) backed by NPMI-scored collocation whitelists. Couldn't beat GPT there because it just "knows" which phrases matter.<p>For the phrase work I had to extract collocations from 100M+ OpenSubtitles lines. Published them as a free dataset: <a href="https://huggingface.co/datasets/vladvlasov256/opensubs-collocations" rel="nofollow">https://huggingface.co/datasets/vladvlasov256/opensubs-collo...</a> There are 43K bigrams across English, Dutch, and Serbian.<p>Source <a href="https://github.com/vladvlasov256/vocab-nlp" rel="nofollow">https://github.com/vladvlasov256/vocab-nlp</a>
The project demonstrates strong technical execution by using classical NLP to solve specific relevance issues in LLM-generated content, but it currently functions as a utility feature rather than a scalable business. While the founder shows impressive data engineering skills, the lack of a defensive moat and the high risk of being 'Sherlocked' by larger language learning platforms or improved LLM prompting limit its venture potential.