閱讀全文 | |
篇名 |
Research of Art Point of Interest Recommendation Algorithm Based on Modified VGG-16 Network
|
---|---|
並列篇名 | Research of Art Point of Interest Recommendation Algorithm Based on Modified VGG-16 Network |
作者 | 劉怡 |
英文摘要 | Traditional point of interest (POI) recommendation algorithms ignore the semantic context of comment information. Integrating convolutional neural networks into recommendation systems has become one of the hotspots in art POI recommendation research area. To solve the above problems, this paper proposes a new art POI recommendation model based on improved VGG-16. Based on the original VGG-16, the improved VGG-16 method optimizes the fully connection layer and uses transfer learning to share the weight parameters of each layer in VGG-16 pre-training model for subsequent training. The new model fuses the review information and user check-in information to improve the performance of POI recommendation. Experiments on real check-in data sets show that the proposed model has better recommendation performance than other advanced points of interest recommendation methods.
|
起訖頁 | 071-085 |
關鍵詞 | art point of interest、VGG-16、transfer learning、recommendation algorithm |
刊名 | 電腦學刊 |
期數 | 202202 (33:1期) |
DOI |
|
QR Code | |
該期刊 上一篇
| Sports Motion Feature Extraction and Recognition Based on a Modified Histogram of Oriented Gradients with Speeded Up Robust Features |
該期刊 下一篇
| A Progressive Real-time Visualization Method for Earthquake Big Data |