PARTNER
๊ฒ€์ฆ๋œ ํŒŒํŠธ๋„ˆ ์ œํœด์‚ฌ ์ž๋ฃŒ

Effective Recommendation Algorithms for Higher Quality Prediction in Collaborative Filtering

ํ•œ๊ตญํ•™์ˆ ์งฟ’์—์„œ ์ œ๊ณตํ•˜๋Š” ๊ตญ๋‚ด ์ตœ๊ณ  ์ˆ˜์ค€์˜ ํ•™์ˆ  ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค๋ฅผ ํ†ตํ•ด ๋‹ค์–‘ํ•œ ๋…ผ๋ๅฉ๊ณผ ํ•™์ˆ ์ง€ ์ •๋ณด๋ฅผ ๋งŒ๋‚˜๋ณด์„ธ์š”.
5 ํŽ˜์ด์งฟ’
๊ธฐํƒ€ํŒŒ์ผ
์ตœ์ดˆ๋“ฑ๋ก์ผ 2025.03.19 ์ตœ์ข…์ ฟ’์ž‘์ผ 2010.11
5P ๋ฏธ๋้ฉ๋ณด๊ธฐ
Effective Recommendation Algorithms for Higher Quality Prediction in Collaborative Filtering
  • ๋ฏธ๋้ฉ๋ณด๊ธฐ

    ์„œ์่ง์ •๋ต–

    ยท ๋ฐœํ–‰๊ธฐ๊ด€ : ํ•œ๊ตญ์ •๋ณด๊ณผํ•™ํšŒ
    ยท ์ˆ˜๋ก์ง€ ์ •๋ณด : ์ •๋ณด๊ณผํ•™ํšŒ ์ปดํ“จํŒ…์˜ ์‹ค์ œ ๋…ผ๋ๅฉ์ง€ / 16๊ถŒ / 11ํ˜ธ / 1116 ~ 1120ํŽ˜์ด์งฟ’
    ยท ์ €์ž๋ช… : ๊น€ํƒํ—Œ, ๋ฐ•์„์ธ, ์–‘์„ฑ๋ด‰

    ์ดˆ๋ก

    In this paper we present two refined neighbor selection algorithms for recommender systems and also show how the attributes of the items can be used for higher prediction quality. The refined neighbor selection algorithms adopt the transitivity-based neighbor selection method using virtual neighbors and alternate neighbors, respectively. The experimental results show that the recommender systems with the proposed algorithms outperform other systems and they can overcome the large scale dataset problem as well as the first rater problem without deteriorating prediction quality.

    ์ฐธ๊ณ ์ž๋ฃŒ

    ยท ์—†์Œ
  • ์ž์ฃผ๋ฌป๋Š”์งˆ๋ๅฉ์˜ ๋‹ต๋ณ€์„ ํ™•์ธํ•ด ์ฃผ์„ธ์š”

    ํ•ดํ”ผ์บ ํผ์Šค FAQ ๋”๋ณด๊ธฐ

    ๊ผญ ์•Œ์•„์ฃผ์„ธ์š”

    • ์ž๋ฃŒ์˜ ์ •๋ณด ๋ฐ ๋‚ด์šฉ์˜ ์ง„์‹ค์„ฑ์— ๋Œ€ํ•˜์—ฌ ํ•ดํ”ผ์บ ํผ์Šค๋Š” ๋ณด์ฆํ•˜์ง€ ์•Š์œผ๋ฉฐ, ํ•ด๋‹น ์ •๋ณด ๋ฐ ๊ฒŒ์‹œ๋ฌผ ์ €์ž‘๊ถŒ๊ณผ ๊ธฐํƒ€ ๋ฒ•์  ์ฑ…์ž„์€ ์ž๋ฃŒ ๋“ฑ๋ก์ž์—๊ฒŒ ์žˆ์Šต๋‹ˆ๋‹ค.
      ์ž๋ฃŒ ๋ฐ ๊ฒŒ์‹œ๋ฌผ ๋‚ด์šฉ์˜ ๋ถˆ๋ฒ•์  ์ด์šฉ, ๋ฌด๋‹จ ์ „์žฌโˆ™๋ฐฐํฌ๋Š” ๊ธˆ์ง€๋˜์–ด ์žˆ์Šต๋‹ˆ๋‹ค.
      ์ €์ž‘๊ถŒ์นจํ•ด, ๋ช…์˜ˆํ›ผ์† ๋“ฑ ๋ถ„์Ÿ ์š”์†Œ ๋ฐœ๊ฒฌ ์‹œ ๊ณ ๊ฐ๋น„๋ฐ”์นด์ง€๋…ธ Viva์˜ ์ €์ž‘๊ถŒ์นจํ•ด ์‹ ๊ณ ๋น„๋ฐ”์นด์ง€๋…ธ Viva๋ฅผ ์ด์šฉํ•ด ์ฃผ์‹œ๊ธฐ ๋ฐ”๋ž๋‹ˆ๋‹ค.
    • ํ•ดํ”ผ์บ ํผ์Šค๋Š” ๊ตฌ๋งค์ž์™ฟ’ ํŒ๋งค์ž ๋ชจ๋‘๊ฐ€ ๋งŒ์กฑํ•˜๋Š” ์„œ๋น„์Šค๊ฐ€ ๋˜๋„๋ก ๋…ธ๋ ฅํ•˜๊ณ  ์žˆ์œผ๋ฉฐ, ์•„๋ž˜์˜ 4๊ฐ€์ง€ ์ž๋ฃŒํ™˜๋ถˆ ์กฐ๊ฑด์„ ๊ผญ ํ™•์ธํ•ด์ฃผ์‹œ๊ธฐ ๋ฐ”๋ž๋‹ˆ๋‹ค.
      ํŒŒ์ผ์˜ค๋ฅ˜ ์ค‘๋ณต์ž๋ฃŒ ์ €์ž‘๊ถŒ ์—†์Œ ์„ค๋ช…๊ณผ ์‹ค์ œ ๋‚ด์šฉ ๋ถˆ์ผ์น˜
      ํŒŒ์ผ์˜ ๋‹ค์šด๋กœ๋“œ๊ฐ€ ์ œ๋Œ€๋กœ ๋˜์ง€ ์•Š๊ฑฐ๋‚˜ ํŒŒ์ผํ˜•์‹์— ๋งž๋Š” ํ”„๋กœ๊ทธ๋žจ์œผ๋กœ ์ •์ƒ ์ž‘๋™ํ•˜์ง€ ์•Š๋Š” ๊ฒฝ์šฐ ๋‹ค๋ฅธ ์ž๋ฃŒ์™ฟ’ 70% ์ด์ƒ ๋‚ด์šฉ์ด ์ผ์น˜ํ•˜๋Š” ๊ฒฝ์šฐ (์ค‘๋ณต์ž„์„ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋Š” ๊ทผ๊ฑฐ ํ•„์š”ํ•จ) ์ธํ„ฐ๋„ท์˜ ๋‹ค๋ฅธ ์‚ฌ์ดํŠธ, ์—ฐ๊ตฌ๊ธฐ๊ด€, ํ•™๊ป“, ์„œ์  ๋“ฑ์˜ ์ž๋ฃŒ๋ฅผ ๋„์šฉํ•œ ๊ฒฝ์šฐ ์ž๋ฃŒ์˜ ์„ค๋ช…๊ณผ ์‹ค์ œ ์ž๋ฃŒ์˜ ๋‚ด์šฉ์ด ์ผ์น˜ํ•˜์ง€ ์•Š๋Š” ๊ฒฝ์šฐ

โ€œ์ •๋ณด๊ณผํ•™ํšŒ ์ปดํ“จํŒ…์˜ ์‹ค์ œ ๋…ผ๋ๅฉ์ง€โ€์˜ ๋‹ค๋ฅธ ๋…ผ๋ๅฉ๋„ ํ™•์ธํ•ด ๋ณด์„ธ์š”!

๋ฌธ์„œ ์ดˆ์•ˆ์„ ์ƒ์„ฑํ•ด์ฃผ๋Š” EasyAI
์•ˆ๋…•ํ•˜์„ธ์š”. ํ•ดํ”ผ์บ ํผ์Šค์˜ ๋ฐฉ๋Œ€ํ•œ ์ž๋ฃŒ ์ค‘์—์„œ ์„ ๋ณ„ํ•˜์—ฌ ๋‹น์‹ ๋งŒ์˜ ์ดˆ์•ˆ์„ ๋งŒ๋“ค์–ด์ฃผ๋Š” EasyAI ์ž…๋‹ˆ๋‹ค.
์ €๋Š” ์•„๋ž˜์™ฟ’ ๊ฐ™์ด ์ž‘์—…์„ ๋„์™ฟ’๋“œ๋ฆฝ๋‹ˆ๋‹ค.
- ์ฃผ์ œ๋งŒ ์ž…๋ ฅํ•˜๋ฉด ๋ชฉ์ฐจ๋ถ€ํ„ฐ ๋ณธ๋ฌธ๋‚ด์šฉ๊นŒ์ง€ ์ž๋™ ์ƒ์„ฑํ•ด ๋“œ๋ฆฝ๋‹ˆ๋‹ค.
- ์žฅ๋ฌธ์˜ ์ฝ˜ํ…์ธ ๋ฅผ ์‰ฝ๊ณ  ๋น ๋ฅด๊ฒŒ ์ž‘์„ฑํ•ด ๋“œ๋ฆฝ๋‹ˆ๋‹ค.
- ์Šคํ† ์–ด์—์„œ ๋ฌด๋ฃŒ ์บ์‹œ๋ฅผ ๊ณ„์ •๋ณ„๋กœ 1ํšŒ ๋ฐœ๊ธ‰ ๋ฐ›์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ง€๊ธˆ ๋ฐ”๋กœ ์ฒดํ—˜ํ•ด ๋ณด์„ธ์š”!
์ด๋Ÿฐ ์ฃผ์ œ๋“ค์„ ์ž…๋ ฅํ•ด ๋ณด์„ธ์š”.
- ์œ ์•„์—๊ฒŒ ์ ํ•ฉํ•œ ๋ฌธํ•™์ž‘ํ’ˆ์˜ ๊ธฐ์ค€๊ณผ ํŠน์„ฑ
- ํ•œ๊ตญ์ธ์˜ ๊ฐ€์น˜๊ด€ ์ค‘์—์„œ ์ •์‹ ์  ๊ฐ€์น˜๊ด€์„ ์ด๋ฃจ๋Š” ๊ฒƒ๋“ค์„ ๋ฌธํ™”์  ๋ฌธ๋ฒ•์œผ๋กœ ์ •๋ฆฌํ•˜๊ณ , ํ˜„๋Œ€ํ•œ๊ตญ์‚ฌํšŒ์—์„œ ์ผ์–ด๋‚˜๋Š” ์‚ฌ๊ฑด๊ณผ ์‚ฌ๊ณ ๋ฅผ ๋น„๊ตํ•˜์—ฌ ์ž์‹ ์˜ ์˜๊ฒฌ์œผ๋กœ ๊ธฐ์ˆ ํ•˜์„ธ์š”
- ์ž‘๋ณ„์ธ์‚ฌ ๋…ํ›„๊ฐ
ํ•ด์บ  AI ์ฑ—๋ด‡๊ณผ ๋Œ€ํ™”ํ•˜๊ธฐ
์ฑ—๋ด‡์œผ๋กœ ๊ฐ„ํŽธํ•˜๊ฒŒ ์ƒ๋‹ดํ•ด๋ณด์„ธ์š”.
2025๋…„ 06์›” 08์ผ ์ผ์š”์ผ
AI ์ฑ—๋ด‡
์•ˆ๋…•ํ•˜์„ธ์š”. ํ•ดํ”ผ์บ ํผ์Šค AI ์ฑ—๋ด‡์ž…๋‹ˆ๋‹ค. ๋ฌด์—‡์ด ๊ถ๊ธˆํ•˜์‹ ๊ฐ€์š”?
9:57 ์˜ค์ „