How do I approach Machine Learning interview questions?

Machine Learning questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master machine learning interviews.

What difficulty level is this interview question?

This is a hard difficulty Machine Learning question, commonly asked during Onsite rounds.

What role is this question designed for?

This question is commonly asked for Data Scientist candidates during technical interviews.

Extract companies from noisy text

Last updated: Jul 4, 2026

Quick Overview

This question evaluates named entity recognition, noisy-text preprocessing, entity disambiguation, and hybrid rule- and model-based pipeline design for extracting organization names from unstructured resumes and web snippets.

|Home/Machine Learning

Extract companies from noisy text

Oct 13, 2025, 9:49 PM

hardData ScientistOnsiteMachine Learning

Extracting Company Names from Noisy Resumes and Web Snippets

Context

You receive messy resume text (PDF-to-text/OCR, varying casing) and scraped web snippets (boilerplate, menus, ads). Your goal is to extract company names (organizations) accurately under noise such as Unicode artifacts, misspellings, acronyms, and ambiguous tokens (e.g., Apple vs apple).

Tasks

(a) Design a hybrid system that combines rule-based patterns (e.g., legal suffixes and context windows) with a machine-learned NER model. Describe the end-to-end pipeline and how you will handle casing, Unicode noise, and misspellings.

(b) Explain feature choices or embeddings (e.g., subword, contextual) and how to incorporate a company gazetteer with fuzzy matching while avoiding label leakage.

(c) Define evaluation metrics (entity-level precision, recall, F1) and an error analysis plan, with special attention to acronyms and ambiguous tokens.

Loading comments...

Browse More Questions

More Machine Learning•More Data Scientist•Data Scientist Machine Learning