Recommender system has become an important component in modern eCommerce. Recent research on recommender systems has been mainly concentrating on improving the relevance or profitability of individual recommended items. But in reality, users are usually exposed to a set of items and they may buy multiple items in one single order. Thus, the relevance or profitability of one item may actually depend on the other items in the set. In other words, the set of recommendations is a bundle with items interacting with each other. In this paper, we introduce a novel problem called the Bundle Recommendation Problem (BRP). By solving the BRP, we are able to find the optimal bundle of items to recommend with respect to preferred business objective. However, BRP is a large-scale NP-hard problem. We then show that it may be sufficient to solve a significantly smaller version of BRP depending on properties of input data. This allows us to solve BRP in real-world applications with millions of users and items. Both offline and online experimental results on a Walmart.com demonstrate the incremental value of solving BRP across multiple baseline models.

No abstract

Existing population-based Stochastic Search Algorithms (SSAs) are too time-consuming to solve dynamic optimal power flow (OPF). The solution proposed in this paper is to accelerate SSAs with memory. Two memory schemes, the similarity retrieval scheme and the mean-based immigrants scheme, are proposed and applied together to the Differential Evolution and Particle Swarm Optimizer, which are two representatives of SSAs. Experiments are conducted on modified IEEE 30-bus and IEEE 118-bus systems with changing load buses and the objective of minimizing real power transmission loss. The results show that the proposed schemes significantly improve the performance of the two existing algorithms, and that SSAs could be practical for tracking optima of dynamic OPF.Index Terms-Dynamic optimal power flow, memory, reactive dispatch problem, stochastic search methods.

''Data sparseness'' is a key issue in current research works on recommendation systems. However, additional information, such as texts, images, knowledge graph, and audios, that is correlated to items helps alleviate the problem to some extent. We focus our research on designing a novel hybrid recommendation system for tourist spots. Tourist spot images are utilized to suppress the ''data sparseness'' problem in the recommendation procedure. First, a novel multimodal visual bayesian personalized ranking algorithm is proposed to fully utilize the cross-modal semantic correlations among different image features. Then, a new recommendation list called L A is generated accordingly from the multimodal perspective. Second, user preference is acquired using the hierarchical sampling statistics model. A new recommendation list called L H is generated in turn from the statistical perspective. Finally, hybrid recommendation results are obtained on the basis of L H and L A. Experimental results demonstrate that the proposed hybrid recommendation system for tourist spots is effective and robust. It is superior to other competitive baselines. More importantly, the proposed hybrid recommendation system is good at recommending a group of tourist spots and more stable than baselines, indicating its high practical value. INDEX TERMS Recommendation system, tourist spots, hierarchical sampling statistics, multimodal visual Bayesian personalized ranking, data sparseness.

No abstract