Abstract ：

为提取电力负荷数据的有效信息，改善传统聚类算法在电力负荷数据中相似度衡量方式单一及聚类效果较差的问题，提出一种采用欧式形态距离的负荷曲线AP聚类方法。首先，使用五分位法将用电负荷曲线重表达为曲线形态变化特征序列，同时使用改进最长公共子序列算法衡量不同特征序列之间的模式匹配度作为曲线之间的差异度。然后，构造一种兼顾曲线整体分布特征和曲线形态变化特征的双尺度相似性度量方法，同时使用熵权法对两种特征进行自适应配比。最后，将所提相似度衡量方法应用到AP(Affinity Propagation)聚类算法中，改进相似度矩阵计算方案，对用户典型日用电负荷曲线进行聚类。算例在标准合成时间序列数据集上进行了实...

Keyword ：

负荷聚类 曲线形态变化特征 双尺度相似性度量 五分位法 智能电网

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Video summarization (VS) is a widely used technique for facilitating the effective reading, fast comprehension, and effective retrieval of video content. Certain properties of the new video data, such as a lack of prominent emphasis and a fuzzy theme development border, disturb the original thinking mode based on video feature information. Moreover, it introduces new challenges to the extraction of video depth and breadth features. In addition, the diversity of user requirements creates additional complications for more accurate keyframe screening issues. To overcome these challenges, this paper proposes a hierarchical spatial–temporal cross-attention scheme for video summarization based on comparative learning. Graph attention networks (GAT) and the multi-head convolutional attention cell are used to extract local and depth features, while the GAT-adjusted bidirection ConvLSTM (DB-ConvLSTM) is used to extract global and breadth features. Furthermore, a spatial–temporal cross-attention-based ConvLSTM is developed for merging hierarchical characteristics and achieving more accurate screening in similar keyframes clusters. Verification experiments and comparative analysis demonstrate that our method outperforms state-of-the-art methods. © 2022 by the authors.

Keyword ：

cross-attention; spatial–temporal features; video summarization

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Multiaccess edge computing (MEC) is a new paradigm to meet the demand of resource-hungry and latency-sensitive services by enabling the placement of services and execution of computing tasks at the edge of radio access networks much closer to resource-constrained devices. However, how to serve more requests while reducing service latency by exploiting limited resources (storage capacities, CPU cycles, communication bandwidth) is still a critical issue in the multidevice MEC-assisted IoT networks, since the time-varying computing demands of devices and unavailability of future information make it difficult to determine where to handle computation tasks and which services to cache. In this article, we propose a twin-timescale framework to jointly optimize adaptive request scheduling (RS) and cooperative service caching (SC) in the multidevices and MEC-assisted networks, in order to explore request dynamic, MECs heterogeneity, service difference. To accommodate the unavailability of future information and unknown system dynamics, we, respectively, formulate RS and SC as partially observable Markov decision process (POMDP) problems. Then, we propose a deep reinforcement learning (DRL)-based online algorithm to improve the service latency reduction ratio and hit rate, which do not require a priori knowledge such as service popularity. Moreover, we give the optimal CPU cycles and communication bandwidth allocations in order to further minimize the average service latency. Extensive and trace-driven simulation results demonstrate the efficacy of the proposed approach. © 2014 IEEE.

Keyword ：

Deep reinforcement learning (DRL); multiaccess edge computing (MEC); request scheduling (RS); resource allocation; service caching (SC)

Cite：

Copy from the list or Export to your reference management。

Abstract ：

With the rapid development of wearable devices with various sensors, massive sensing data for health management have been generated. This causes a potential revolution in medical treatments, diagnosis, and prediction. However, due to the privacy risks of health data aggregation, data comparative analysis under privacy protection faces challenges. Order-preserving encryption is an effective scheme to achieve private data retrieval and comparison, but the existing order-preserving encryption algorithms are mainly aimed at either integer data or single characters. It is urgent to build a lightweight order-preserving encryption scheme that supports multiple types of data such as integer, floating number, and string. In view of the above problems, this paper proposes an order-preserving encryption scheme (WRID-OPES) based on weighted random interval division (WRID). WRID-OPES converts all kinds of data into hexadecimal number strings and calculates the frequency and weight of each hexadecimal number. The plaintext digital string is blocked and recombined, and each block is encrypted using WRID algorithm according to the weight of each hexadecimal digit. Our schemes can realize the order-preserving encryption of multiple types of data and achieve indistinguishability under ordered selection plaintext attack (IND-OCPA) security in static data sets. Security analysis and experiments show that our scheme can resist attacks using exhaustive methods and statistical methods and has linear encryption time and small ciphertext expansion rate.

Keyword ：

IoT; order-preserving encryption; privacy protection; random interval division; wearable devices

Cite：

Copy from the list or Export to your reference management。

Abstract ：

In recent years, with the popularity of smartphones, the indoor positioning systems based on mobile crowdsensing (MCS) have gained considerable interest and exploit. However, it is still challenging to construct a largescale indoor positioning system. 1) In indoor positioning model, storage and computing resources are very important. 2) The calibration operation of data label and selection of model parameters require the operation of professionals. 3) User location privacy may be compromise, which greatly affects participant safety and enthusiasm. To solve these problems, our model firstly provides an edge-crowdsourcing indoor localization architecture to improve storage, computing power and response speed. Then, based on manifold regularization, a semi-supervised indoor localization model is determined by an adaptive manner in terms of both similarity and manifold structure, which reduces the workload of the positioning model and improve localization accuracy. In addition, we propose a new privacy-aware indoor localization algorithm based on secure multi-party computation to protect location privacy. Experimental results on real-world datasets show that, compared with the previous methods, our method improves accuracy by 0.87m, and in terms of time overhead of privacy protection, our method reduces the running time of the thousand seconds level. © 2021 IEEE.

Keyword ：

Digital storage Indoor positioning systems Privacy by design Storage as a service (STaaS)

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Dialogue-based multiple-choice machine reading comprehension (MRC) is one of most difficult and novel tasks because it requires more advanced reading comprehension skills, such as speaker's intention analysis, non-extractive reasoning, commonsense knowledge. Previous models usually only compute attention scores from the fixed representation of entire dialogue, and also do not fully consider the contribution of dialogue, question, options, and their combinations respectively. In this paper, we introduce Utterance-focusing Multiway-matching Network (UMN), a simple but effective human mimicking model for dialogue-based multiple-choice MRC. First, two utterance-focusing mechanisms called ParaUF and AutoUF are proposed to extract the utterances that are most relevant to the question and option: ParaUF gets the bilinear weighted distance between each utterance of dialogue and question and option during training while AutoUF obtains the scores by the relevance, overlap and coverage (ROC) rules before training process. Second, we adopted the multiway-matching mechanism to capture the relationship among the question, option and selected utterances through calculating the attention weights between the quadruplet of four sequences: utterances, question, option and the concatenation of each two. We evaluate the proposed model on dialogue-based multiple-choice MRC tasks, DREAM, and outperformed recently published methods under the same pre-trained model. A series of detailed analysis is also conducted to interpret the differences of two utterance-focusing mechanisms and the effectiveness of the proposed multiway-matching mechanism. © 2020 Elsevier B.V.

Keyword ：

Computer applications Focusing Neural networks

Cite：

Copy from the list or Export to your reference management。

Abstract ：

Computation offloading has been considered as a promising solution for resource-constrained mobile devices supporting computation-intensive mobile collaborative applications in beyond 5G networks. By leveraging network-assisted device-to-device (D2D) collaboration, resource-constrained mobile devices are able to offload computation tasks to nearby resource-sharing mobile devices. Meanwhile, blockchain technology is developing rapidly and has been applied to mobile scenarios with massive information interaction for establishing trust between mobile devices. However, the computing power of mobile devices is limited by the physical size and battery capacity, which is not enough to cope with the high computation overhead for blockchain mining process. Thus, offloading mining tasks to the edge servers becomes a viable option. This paper investigates both of the D2D task offloading problem and the mining task offloading problem in a blockchain-enabled beyond 5G network. For the purpose of efficient resource allocation, the former is formulated as a double auction market and the latter is formulated as a Stackelberg game. Moreover, we propose two pricing based schemes to solve these two computation offloading problems, among which Bayes-Nash equilibrium and Stackelberg equilibrium are analyzed. At last, by comparison with the benchmark, the efficiency and effectiveness of our proposed schemes are evaluated. Numerical results show that both of the double auction based normal task offloading scheme and the Stackelberg game based mining task offloading scheme are able to improve the utility of resource demanders and providers. © 2021

Keyword ：

5G mobile communication systems Blockchain Computation theory Computer games Costs Economics Game theory Mobile edge computing Queueing networks Resource allocation

Cite：

Copy from the list or Export to your reference management。

Abstract ：

机器阅读理解的目标是使机器理解自然语言文本,并能够正确回答与文本相关的问题.由于数据集规模的制约,早期的机器阅读理解方法大多基于人工特征以及传统机器学习方法进行建模.近年来,随着知识库、众包群智的发展,研究者们陆续提出了高质量的大规模数据集,为神经网络模型以及机器阅读理解的发展带来了新的契机.对基于神经网络的机器阅读理解相关的最新研究成果进行了详尽的归纳:首先,概述了机器阅读理解的发展历程、问题描述以及评价指标;然后,针对当前最流行的神经阅读理解模型架构,包括嵌入层、编码层、交互层和输出层中所使用的相关技术进行了全面的综述,同时阐述了最新的BERT预训练模型及其优势;之后,归纳了近年来机器阅读理解数据集和神经阅读理解模型的研究进展,同时,详细比较分析了最具代表性的数据集以及神经网络模型;最后展望了机器阅读理解研究所面临的挑战和未来的研究方向.

Keyword ：

机器阅读理解 神经阅读理解模型 注意力机制 自然语言处理

Cite：

Copy from the list or Export to your reference management。

Abstract ：

针对已有大多数研究在设计激励机制时未考虑用户的隐私泄露问题,本文提出一种支持隐私保护的激励机制综合方案IMPP(Incentive Mechanism with Privacy-Preserving in mobile crowd sensing).首先,基于轻量级隐私保护思想,采用单向安全哈希函数生成256位哈希值作为参与者的匿名身份标识,以此来保护参与者的身份隐私;其次,依据参与者的数据效用值、期望回报及感知任务预算实现面向数据质量的补偿激励,选择性价比最高的胜出者;接着,借助分布式压缩感知理论,对胜出者的原始感知数据压缩处理,得到剔除冗余的观测值,并在观测值中添加哈希函数值等噪扰数据后传送于服务器端聚合,以增强感知数据的隐私性保护,之后对隐私数据集进行完整性校验并重构;最后,利用真实数据集,通过仿真实验对IMPP的有效性进行对比分析.实验结果表明,IMPP机制在隐私保护水平、数据完整性、数据精确性、时间效率、评估准确率、重构匹配度及激励效果等方面是高效的.

Keyword ：

分布式压缩感知 哈希函数 激励机制 群智感知 隐私保护

Cite：

Copy from the list or Export to your reference management。

Abstract ：

针对已有大多数研究在设计激励机制时未考虑用户的隐私泄露问题,本文提出一种支持隐私保护的激励机制综合方案IMPP(Incentive Mechanism with Privacy-Preserving in mobile crowd sensing).首先,基于轻量级隐私保护思想,采用单向安全哈希函数生成256位哈希值作为参与者的匿名身份标识,以此来保护参与者的身份隐私;其次,依据参与者的数据效用值、期望回报及感知任务预算实现面向数据质量的补偿激励,选择性价比最高的胜出者;接着,借助分布式压缩感知理论,对胜出者的原始感知数据压缩处理,得到剔除冗余的观测值,并在观测值中添加哈希函数值等噪扰数据后传...

Keyword ：

分布式压缩感知 哈希函数 激励机制 群智感知 隐私保护

Cite：

Copy from the list or Export to your reference management。