Buyer intent in e-commerce is multi-faceted and is expressed through explicit attributes—such as brand, size, color, and material, rather than through general topical relevance. However, many state-of-the-art scalable query-product matching systems rely on aggregate representations, scoring a single query embedding against a single item embedding. While efficient, this aggregation frequently fails to satisfy individual attribute intent: items can be semantically related, yet violate key aspects specified in the query. In contrast, fine-grained interaction methods can better capture aspect-level constraints, but are typically too expensive due to increased run-time computation and storage costs. We propose an aspect-aware ranking framework that retrieves and resolves aspects in queries and performs fine-grained semantic affinity match against aspects in products to compute an aggregate query-product level aspect affinity score. The proposed approach integrates (i) query aspect resolution (canonicalization) using structured aspect data, (ii) a model to learn granular aspect affinity signal capturing individual aspect-level understanding; and iii) an efficient design for online serving, significantly cutting cost associated with inference speed and storage. This design preserves the scalability of two-tower retrieval while substantially improving explicit intent satisfaction.

With the vigorous development of the Internet of Things, the Internet, cloud computing, and mobile terminals, edge computing has emerged as a new type of Internet of Things technology, which is one of the important components of the Industrial Internet of Things. In the face of large-scale data processing and calculations, traditional cloud computing is facing tremendous pressure, and the demand for new low-latency computing technologies is imminent. As a supplementary expansion of cloud computing technology, mobile edge computing will sink the computing power from the previous cloud to a network edge node. Through the mutual cooperation between computing nodes, the number of nodes that can be calculated is more, the types are more comprehensive, and the computing range is even greater. Broadly, it makes up for the shortcomings of cloud computing technology. Although edge computing technology has many advantages and has certain research and application results, how to allocate a large number of computing tasks and computing resources to computing nodes and how to schedule computing tasks at edge nodes are still challenges for edge computing. In view of the problems encountered by edge computing technology in resource allocation and task scheduling, this paper designs a dynamic task scheduling strategy for edge computing with delay-aware characteristics, which realizes the reasonable utilization of computing resources and is required for edge computing systems. This paper proposes a resource allocation scheme combined with the simulated annealing algorithm, which minimizes the overall performance loss of the system while keeping the system low delay. Finally, it is verified through experiments that the task scheduling and resource allocation methods proposed in this paper can significantly reduce the response delay of the application.

Entity Set Expansion (ESE) and Attribute Extraction (AE) are usually treated as two separate tasks in Information Extraction (IE). However, the two tasks are tightly coupled, and each task can benefit significantly from the other by leveraging the inherent relationship between entities and attributes. That is, 1) an attribute is important if it is shared by many typical entities of a class; 2) an entity is typical if it owns many important attributes of a class. Based on this observation, we propose a joint model for ESE and AE, which models the inherent relationship between entities and attributes as a graph. Then a graph reinforcement algorithm is proposed to jointly mine entities and attributes of a specific class. Experimental results demonstrate the superiority of our method for discovering both new entities and new attributes.

Summary Many high performance computing applications require computing both sparse matrix‐vector product (SMVP) and sparse matrix‐transpose vector product (SMTVP) for better overall performance. Under such a circumstance, it is critical to maintain a similarly high throughput for these two computing patterns with the underlying sparse matrix encoded in a single storage format. The compressed sparse block (CSB) format proposed by Buluç et al. allows computing both problems on multi‐core CPUs with nearly identical throughputs. On the other hand, a direct porting of CSB to graphics processing units (GPUs), which have been recently recognized as a powerful general purpose computing platform, turns out to be inefficient. In this work, we propose a new data structure, designated as expanded CSB (eCSB), to minimize the throughput gap between SMVP and SMTVP computations on GPUs, while at the same time enable a high computing throughput. We also use a hybrid storage format to store elements in each block, which can be selected dynamically at runtime. Experimental results show that the proposed techniques implemented on a Kepler GPU delivers similar throughput on both SMVP and SMTVP and the throughput is up to 13 times faster than that of the CPU‐based CSB implementation. In addition, our eCSB procedure outperforms the previous GPU results by up to 188% and 914% in computing SMVP and SMTVP, and we validate the effectiveness of eCSB by means of wall‐clock time of bi‐conjugate gradient algorithm; our eCSB is 25% faster than Compressed Sparse Rows (CSR) and 6% faster than HYB, respectively. Copyright © 2014 John Wiley & Sons, Ltd.

To accomplish a search task and satisfy a single information need, users usually submit a series of queries to web search engines. It is useful for web search engines to detect the task boundaries in a series of successive queries. Traditional task boundary detection methods are based on time gap and lexical comparisons, which often suffer from the vocabulary gap problem, that is, the topically related queries may not share any common words. In this paper we learn hidden topics from query log and leverage them to resolve the vocabulary gap problem. Unlike other external knowledge resources, such as WordNet and Wikipedia, the hidden topics discovered from query log cover long tail queries, which is useful to detect task boundaries. Experimental results on dataset from real world query log demonstrate that the proposed method achieves significant quality enhancement.

Infrastructure as a Service (IaaS) is important in Cloud Computing, which provides on-demand virtual machines (VMs) to users. Proper deployment of the virtual machine onto available hosts plays an important role on the load-balancing of modern data center. At present, many load-balancing methods are based on the load forecasting model, that could predict the resource requirements of the virtual machine. However, the resource requirement of virtual machine in IaaS Cloud is hard to predict because there will be variety of load types in IaaS cloud. Moreover, a variety of heterogeneous hardware environments and virtualization technologies make it hard to predict the requirement of virtual machine based on the workloads. To address the problem, we propose a model based method to predict and calculate the resource requirement of each virtual machine, and using this model to design a load-balancing framework in IaaS Cloud. The contribution of this paper includes: (1) A model that forecast the load and estimate the resource requirement of virtual machines in IaaS Cloud, (2) A scalable framework for load-balancing which uses our resource requirement forecasting model. Experiments show that our method can accurately estimate the resource requirements of virtual machines, and work well in our load-balancing framework.

Virtualization technology is one of the key technologies in cloud computing. Proper allocation of the virtual machines onto available hosts plays an important role on the performance optimization of cloud computing. At present, many resource management methods are based on the load model to predict the resource requirements of the application. However, a variety of heterogeneous hardware environments and virtualization technologies make it hard to efficiently allocate the resources based on existing methods. To address the problem, we proposed a virtual machine allocation strategy based on statistics. And The contribution of this paper include:(1) A load-resource model for estimating the resource demand of each virtual machine, (2) An algorithm for assigning virtual machines onto the hosts among resource pool according to the resources requirement of virtual machine. Experiments show that our load-resource model can accurately estimate the resource requirements of virtual machines for physical host, and our virtual machine assigning algorithm can achieve better load balancing compared with the first-fit and best-fit algorithm.

The text classification methods heavily depend on machine learning algorithms with abstract mathematic metrics, which obstruct the direct observation and intuitive understanding of the text-specific classification. In this paper, we model a document as a Document-Classes-Topics top-down hierarchical structure. Furthermore, by running the document generation procedure, we can obtain each class's content share, which not only can be used to make the classification decision but also can provide a natural visualization approach for text classification. We implement this idea by a new tool named TC-DCA, which provides the visualization of text classification result, where the target document is expressed graphically as its content's allocation on every class. TC-DCA can also perform the drilling down operation to reveal the classification effect of each word of the document.