Kemiripan pertanyaan dan ketepatan gramatikal

Natural Language Processing (NLP) in Python

Fouad Trad

Machine Learning Engineer

Kemiripan pertanyaan

Mengidentifikasi apakah dua pertanyaan merupakan parafrase
Berguna untuk:
- Deduplikasi
- Klasterisasi pertanyaan serupa
- Meningkatkan akurasi pencarian
Dilatih pada dataset Quora Question Pairs (QQP)

Gambar tiga orang mengajukan pertanyaan.

Pipeline QQP

from transformers import pipeline

qqp_pipeline = pipeline(
    task="text-classification", 
    model="textattack/bert-base-uncased-QQP"
    )

question1 = "How can I learn Python?"
question2 = "What is the best way to study Python?"

result = qqp_pipeline({"text": question1, "text_pair": question2})

print(result)

{'label': 'LABEL_1', 'score': 0.6853412985801697}

Pipeline QQP

from transformers import pipeline
qqp_pipeline = pipeline(
    task="text-classification", 
    model="textattack/bert-base-uncased-QQP"
    )
question1 = "How can I learn Python?"
question2 = "What is the capital of France?"
result = qqp_pipeline({"text": question1, "text_pair": question2})
print(result)

{'label': 'LABEL_0', 'score': 0.9999338388442993}

Menilai ketepatan tata bahasa

Menilai tingkat ketepatan tata bahasa sebuah teks
Berguna untuk:
- Alat edukasi
- Pemeriksa tata bahasa
- Asisten penulisan
Dilatih pada dataset Corpus of Linguistic Acceptability (CoLA)

Gambar seseorang menilai ketepatan teks tertulis.

Pipeline CoLA

from transformers import pipeline
cola_classifier = pipeline(
  task="text-classification", 
  model="textattack/distilbert-base-uncased-CoLA"
)

result = cola_classifier("The cat sat on the mat.")

print(result)

[{'label': 'LABEL_1', 'score': 0.9918296933174133}]

Pipeline CoLA

from transformers import pipeline
cola_classifier = pipeline(
  task="text-classification", 
  model="textattack/distilbert-base-uncased-CoLA"
)
result = cola_classifier("The cat on sat mat the.")
print(result)

[{'label': 'LABEL_0', 'score': 0.9628171324729919}]

Ayo berlatih!

Natural Language Processing (NLP) in Python