In order to solve the problem of huge labeling cost for person re-identification， a method of one-shot video-based person re-identification with multi-loss learning and joint metric was proposed. Aiming at the problem that the number of label samples is small and the model obtained is not robust enough， a Multi-Loss Learning （MLL） strategy was proposed. In each training process， different loss functions were used for different data to optimize and improve the discriminative ability of the model. Secondly， a Joint Distance Metric （JDM） was proposed for label estimation， which combined the sample distance and the nearest neighbor distance to further improve the accuracy of pseudo label prediction. JDM solved the problems of the low accuracy of label estimation for unlabeled data， and the instability in the training process caused by the unlabeled data not fully utilized. Experimental results show that compared with the one-shot progressive learning method PL （Progressive Learning）， the rank-1 accuracy reaches 65.5% and 76.2% on MARS and DukeMTMC-VideoReID datasets when the ratio of pseudo label samples added per iteration is 0.10， with the improvement of the proposed method of 7.6 and 5.2 percentage points， respectively.

Focusing on user remarkable burden and high dimension of Multi-Criteria Collaborative Filtering (MC-CF) recommendation system, the recommendation system of Multi-Attribute Utility Collaborative Filtering (MAU-CF) was proposed. Firstly, attribute weight and attribute-value utility were extracted by user browsing behavior, and user's multi-attribute utility function was built to achieve implicit rating of items. Secondly, attribute-value collection according to user preference was constructed based on Genetic Algorithm (GA). Thirdly, the nearest neighborhood was looked for by attribute weight and attribute-value similarity of attribute-value collection. Finally, utilities of items which the nearest neighborhood had browed and bought would be predicted for user by similarity, and the high-utility items would be recommended to user. In the comparison experiments with MC-CF, the explicit utility was replaced by the implicit utility calculated by MAU-CF, calculation dimension decreased by 44.16%, time expense decreased by 27.36%, and Mean Absolute Error (MAE) decreased by 5.69%, and user satisfaction increased by 13.44%. The experimental results show MAU-CF recommendation system outperforms MC-CF recommendation system on user burden, calculation dimension, and recommendation quality.