篇名 |
Efficient Barrage Video Recommendation Algorithm Based on Convolutional and Recursive Neural Network
|
---|---|
並列篇名 | Efficient Barrage Video Recommendation Algorithm Based on Convolutional and Recursive Neural Network |
作者 | Ping He、Siyuan Ma、Weidong Li |
英文摘要 | Convolutional neural network has been widely used in recommendation methods. Recursive convolutional neural network not only has the advantages of automatic feature extraction, but also effectively captures the features before and after the time series. However, the traditional video recommendation process has the problems of low time efficiency and low accuracy. To solve the above problems, this paper proposes a method that is based on convolutional recurrent neural network in the barrage video recommendation. Firstly, according to the number of barrage videos, the method selects the preferable video fragments of users, and adopts the kmeans clustering method to extract key frames from video. Secondly, we use a convolution and recursive neural network model (RCNN) to classify the similar video fragments. Finally, the recommendation can be achieved by the similarity between video fragments. Experimental results demonstrate the effectiveness and better performance of the proposed method. |
起訖頁 | 1241-1251 |
關鍵詞 | Barrage、Convolutional neural network、Recursive neural network、Video recommendation |
刊名 | 網際網路技術學刊 |
期數 | 202111 (22:6期) |
出版單位 | 台灣學術網路管理委員會 |
DOI |
|
QR Code | |
該期刊 上一篇
| PAGS: PUF-based Anonymous Group Signature for Node Authentication in Edge Computing |
該期刊 下一篇
| Expert System for Drying Schemes of Agricultural Products Based on Fuzzy Control and BP Neural Network |