본문으로 건너뛰기
AI 도감

텍스트 분류와 감정 분석

글에 미리 정한 라벨을 붙인다

언어와 지식입문 #05
입력텍스트텍스트

이 페이지의 본문은 영어로 제공됩니다. 제목과 요약은 한국어로 번역되었습니다.

이 능력이 뜻하는 것

Takes a piece of text and outputs its category — sentiment polarity, topic, intent or policy violation. The label set is fixed at training time, so it answers which class rather than what is said. A single text can carry several labels, and the model can also emit a probability distribution over the label set.

기술적으로 구현하는 방법

Early systems paired bag-of-words with linear classifiers; fine-tuning a pre-trained encoder such as the BERT family later became the standard when labelled data is scarce. A classification head sits on top of the encoder and is trained with cross-entropy. With many labels and few examples, it can be recast as sentence-pair entailment or done zero-shot by prompting a large model.

대표 제품

5

관련 기관

대표적 용도

  • Sentiment polarity of reviews and public opinion
  • Spam and policy-violation filtering
  • Routing tickets by topic and priority
  • Intent detection to drive dialogue

성능을 평가하는 방법

F1
Harmonic mean of precision and recall; more trustworthy than accuracy under imbalance
ROC-AUC
Ranking quality, threshold-independent
Confusion matrix
Per-class error direction, to locate systematic bias

경계와 난점

  • Under heavy class imbalance, minority classes are ignored while accuracy still looks high
  • Wording drift outside the training domain degrades predictions
  • Sarcasm, negation and double negation are often flipped, since the literal words point the other way

뒤에 있는 개념