NLP Applications and Ethics

NLP technology is now embedded in virtually every sector of society — from healthcare and education to social media and law. With this widespread adoption comes significant ethical responsibilities. This lesson surveys the major applications of NLP and examines the ethical challenges that practitioners must address.

---

Major NLP Applications

Machine Translation

Machine translation (MT) automatically translates text between languages.

Generation	Approach	Example
1st generation	Rule-based	Hand-crafted grammar rules
2nd generation	Statistical (SMT)	Phrase-based translation
3rd generation	Neural (NMT)	Seq2Seq with attention
Current	Transformer-based	Google Translate (Transformer), DeepL

from transformers import pipeline

translator = pipeline("translation", model="Helsinki-NLP/opus-mt-en-de")
result = translator("Machine translation has improved dramatically.")
print(result[0]['translation_text'])

Question Answering

Type	Description	Example
Extractive QA	Extracts an answer span from a given passage	SQuAD dataset
Abstractive QA	Generates a new answer (not just a span)	GPT-based QA
Open-domain QA	Finds answers from a large knowledge base	Google Search, Bing Chat

from transformers import pipeline