Extracting Information from Text
🔑 Key techniques
1. Named Entity Recognition (NER)
Extracts people, places, dates, organisations
Example:
“Microsoft was founded by Bill Gates” → Microsoft (Org), Bill Gates (Person)
2. Key Phrase Extraction
Pulls out important terms
Example:
“Fabric simplifies data engineering workflows” → Fabric, data engineering
3. Sentiment Analysis
Detects positive / negative / neutral tone
Used in reviews, feedback
4. Text Classification
Assigns categories
Example: Spam vs Not Spam
5. Summarisation
Produces short summaries of long text
🧠Model choice logic
Scenario Technique
Extract names, dates NER
Understand opinion Sentiment Analysis
Categorise text Classification
Get key topics Key Phrase Extraction
