Rating Prediction Method for Recommendation Algorithm Based on Observed Ratings and Similarity Graphs
The present invention relates to a rating prediction method of a recommendation algorithm using an observation rating and a similarity graph. The rating prediction method according to the present invention can be executed in a recommendation system in various fields, or a rating prediction apparatus...
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Patent |
| Sprache: | eng ; kor |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Zusammenfassung: | The present invention relates to a rating prediction method of a recommendation algorithm using an observation rating and a similarity graph. The rating prediction method according to the present invention can be executed in a recommendation system in various fields, or a rating prediction apparatus provided in the same, and can increase the accuracy of rating prediction and maximize the performance of rating prediction by using a similarity graph additionally given, when provided information is insufficient in addition to the observation rating.
본 발명은 관측평점과 유사도 그래프를 활용한 추천 알고리즘의 평점 예측 방법에 관한 것으로서, 다양한 분야의 추천 시스템, 또는 그에 구비된 평점 예측 장치에서 실행될 수 있는, 본 발명의 평점 예측 방법은, 관측평점 이외에도 관측평점으로 주어진 정보가 부족한 경우에 추가로 주어진 유사도 그래프(Similarity Graph)를 더 이용하여 평점 예측의 정확도를 높이고 평점 예측의 성능을 극대화시키는 추천 알고리즘의 평점 예측 방법을 제공한다. |
|---|