๊ด€๋ฆฌ ๋ฉ”๋‰ด

๋ชฉ๋ก์ด์‚ฐ ํ‘œํ˜„๋ฐฉ๋ฒ• (1)

DATA101

[NLP] Bag of Words(BoW) ๊ฐœ๋… ๋ฐ ์‹ค์Šต

๋ณธ ํฌ์ŠคํŒ…์—์„œ๋Š” ์นด์šดํŠธ ๊ธฐ๋ฐ˜์˜ ๋‹จ์–ด ํ‘œํ˜„ ๋ฐฉ๋ฒ•์ธ Bag of Words(BoW) ๊ฐœ๋…๊ณผ ์ƒ์„ฑ ๋ฐฉ๋ฒ•์„ ์•Œ์•„๋ด…๋‹ˆ๋‹ค.๐Ÿ“š ๋ชฉ์ฐจ1. BoW ๊ฐœ๋…2. BoW ํŠน์ง•3. BoW ์ƒ์„ฑ ์ ˆ์ฐจ4. BoW ์ƒ์„ฑ ์‹ค์Šต1. BoW ๊ฐœ๋…Bag of Words(BoW)๋Š” ๋‹จ์–ด๋ฅผ ์ˆ˜์น˜ํ™”ํ•˜๋Š” ๋ฐฉ๋ฒ• ์ค‘ ํ•˜๋‚˜๋กœ, ๋ฌธ์„œ ๋‚ด ๋‹จ์–ด์˜ ์ˆœ์„œ์™€ ์˜๋ฏธ๋Š” ๊ณ ๋ คํ•˜์ง€ ์•Š๊ณ  ์˜ค์ง ์ถœํ˜„ ๋นˆ๋„(frequency)๋งŒ ๊ณ ๋ คํ•˜์—ฌ ๋‹จ์–ด๋ฅผ ํ‘œํ˜„ํ•˜๋Š” ๋ฐฉ๋ฒ•์ž…๋‹ˆ๋‹ค. BoW๋Š” ๊ตญ์†Œ ํ‘œํ˜„๋ฐฉ๋ฒ•(Local Representation) ๋˜๋Š” ์ด์‚ฐ ํ‘œํ˜„๋ฐฉ๋ฒ•(Discrete Representation)์˜ ์ผ์ข…์œผ๋กœ ์นด์šดํŠธ ๊ธฐ๋ฐ˜์˜ ๋‹จ์–ด ํ‘œํ˜„๋ฐฉ๋ฒ•(Count-based Word Representation)์ด๋ผ๊ณ  ๋ถ€๋ฆ…๋‹ˆ๋‹ค(๊ทธ๋ฆผ 1 ์ฐธ๊ณ ).2. BoW ํŠน์ง•BoW๋Š” ์–ด๋–ค ๋‹จ์–ด๋“ค์ด ๋ช‡ ํšŒ..