πŸ€μ˜€μ›”μ˜ 코딩일지

Artificial Intelligence

(23.03.10)Pythonν”„λ‘œκ·Έλž˜λ°: 인곡신경망을 ν†΅ν•œ μΆ”μ²œ μ‹œμŠ€ν…œ Neural Collaborative Filtering(NCF), ν•¨μˆ˜ν˜• 신경망

ν”„λ‘œκ·Έλž˜λ¨Έ μ˜€μ›” 2023. 3. 23.

인곡신경망을 ν†΅ν•œ μΆ”μ²œ μ‹œμŠ€ν…œ

 

 

- Neural Collaborative Filtering(NCF)

-μˆœμ°¨ν˜• (Squential) 신경망 생성법

-ν•¨μˆ˜ν˜•(Functional) 신경망 생성법

- μ§€κΈˆκΉŒμ§€ λ‚˜μ˜¨ μΆ”μ²œ 방식 μ€‘μ—μ„œ κ°€μž₯ 쒋은 μ„±λŠ₯

- Regression λ°©μ‹μœΌλ‘œ 뢄석가λŠ₯

- μ˜ν™”μ˜ 평점정보 (userid, movieid, rating)

- μ΄μš©μžλŠ” μ˜ν™”μ— λŒ€ν•œ μ·¨ν–₯이 λͺ¨λ‘ λ‹€λ₯΄λ‹€

- μ˜ν™”λŠ” λ‹€μ–‘ν•œ μž₯λ₯΄κ°€ ν˜Όν•©λ˜μ–΄μžˆλ‹€

- μ΄μš©μžλŠ” μžμ‹ μ˜ μ·¨ν–₯에 λ§žλŠ” μ˜ν™”μ— 높은 rating을 μ œμ‹œν•¨

- μ–΄λ–€ μ΄μš©μžμ—κ²Œ μ–΄λ–€μž₯λ₯΄μ˜ μ˜ν™”λ₯Ό μΆ”μ²œν•  것인가?

 

μž„λ² λ”© λ ˆμ΄μ–΄ -λ²”μ£Όν˜•μ„ μˆ˜μΉ˜ν˜•μœΌλ‘œ λ°”κΏ”μ„œ ν‘œν˜„, λ‹¨μˆœν•œ μˆ«μžκ°€ μ•„λ‹Œ λ²‘ν„°λ‘œ λ°”κΏ”μ€€λ‹€.

λ‹¨μˆœν•œ 슀칼라λ₯Ό μ‚¬μš©ν•˜μ§€ μ•ŠλŠ”λ‹€.

 

λ²‘ν„°μ˜ λ°©ν–₯은 μ„±ν–₯으둜 생각할 수 μžˆλ‹€.

μ’‹μ•„ν•˜λŠ” μ·¨ν–₯에 λŒ€ν•œ μˆ«μžλŠ” λŒ€λž΅ 20κ°œκ°€ μ λ‹Ήν•˜λ‹€

20차원 벑터

20가지 μ’…λ₯˜μ˜ μ·¨ν–₯

μ§€λ„ν•™μŠ΅ νšŒκ·€λΆ„μ„

 

INPUT

CALLABLE

 

 

 

 

'Artificial Intelligence' μΉ΄ν…Œκ³ λ¦¬μ˜ λ‹€λ₯Έ κΈ€

(23.03.09.)Python ν”„λ‘œκ·Έλž˜λ°: CNN(Convolutional Neural Network) 이미지 λΆ„λ₯˜ ν•˜κΈ°  (0) 2023.03.23
(23.03.08.)Python ν”„λ‘œκ·Έλž˜λ°: Tensorflow - μ§€λ„ν•™μŠ΅, 닀쀑뢄λ₯˜ 뢄석 (iris(뢓꽃) ν’ˆμ’… λΆ„λ₯˜ 뢄석)  (0) 2023.03.23
(23.03.07)Python ν”„λ‘œκ·Έλž˜λ°: Tensorflow-νšŒκ·€λΆ„μ„, ReLU(Rectified Linear Unit)  (0) 2023.03.23
(23.03.06)Python ν”„λ‘œκ·Έλž˜λ°: ν™œμ„±ν•¨μˆ˜, κ³„λ‹¨ν•¨μˆ˜ - sigmoid  (0) 2023.03.23
(23.02.28)Python ν”„λ‘œκ·Έλž˜λ°: 비지도 ν•™μŠ΅ - DBSCAN(Density-based spatial clustering of applications with noise)(μ—°μ†ν˜•+ λ²”μ£Όν˜•)  (0) 2023.03.06

λŒ“κΈ€

ν‹°μŠ€ν† λ¦¬νˆ΄λ°”