Interview: 'you have 1M support tickets — how do you categorize them?'
hardAnswer
- (1) Deduplicate near-identical tickets (LSH).
- (2) Sentence-transformer embed (bge-small or multilingual if needed).
- (3) UMAP to 20 dim.
- (4) HDBSCAN → topic clusters.
- (5) c-TF-IDF for cluster keywords → auto-label.
- (6) Manual review of top 20 clusters + rename/merge/split with domain team.
- (7) Train classifier on labeled clusters for future ticket routing.
- (8) Monitor cluster drift monthly + retrain HDBSCAN if new topics emerge (novelty detection).
Check yourself — multiple choice
- Random
- Dedup (LSH) + sentence-transformers + UMAP + HDBSCAN + c-TF-IDF keywords + human review + classifier for routing + monthly drift monitoring
- Just LDA
- Not real
1M tickets: dedup + embed + UMAP + HDBSCAN + c-TF-IDF + review + classifier.
#interview#applications#nlp
Practise Unsupervised Learning
214 interview questions in this topic.
Related questions
- Interview: users complain search returns irrelevant results — how do you fix?
- Interview: choose a sentence embedding model for a startup RAG.
- Interview: 'you're asked to segment 10M customers — how?'
- Interview: 'design an unsupervised fraud detection system'.
- BM25 — why is it still competitive with modern retrievers?
- Vector search / approximate nearest neighbors — main algorithms.