An Efficient Modelling of Oversampling with Optimal Deep Learning Enabled Anomaly Detection in Streaming Data

Recently,anomaly detection(AD)in streaming data gained significant attention among research communities due to its applicability in finance,business,healthcare,education,etc.The recent developments of deep learning(DL)models find helpful in the detection and classification of anomalies.This article designs an oversampling with an optimal deep learning-based streaming data classification(OS-ODLSDC)model.The aim of the OSODLSDC model is to recognize and classify the presence of anomalies in the streaming data.The proposed OS-ODLSDC model initially undergoes preprocessing step.Since streaming data is unbalanced,support vector machine(SVM)-Synthetic Minority Over-sampling Technique(SVM-SMOTE)is applied for oversampling process.Besides,the OS-ODLSDC model employs bidirectional long short-term memory(Bi LSTM)for AD and classification.Finally,the root means square propagation(RMSProp)optimizer is applied for optimal hyperparameter tuning of the Bi LSTM model.For ensuring the promising performance of the OS-ODLSDC model,a wide-ranging experimental analysis is performed using three benchmark datasets such as CICIDS 2018,KDD-Cup 1999,and NSL-KDD datasets.

The viscous strip approach to simplify the calculation of the surface acoustic wave generated streaming

In recent decades,the importance of surface acoustic waves,as a biocompatible tool to integrate with microfluidics,has been proven in various medical and biological applications.The numerical modeling of acoustic streaming caused by surface acoustic waves in microchannels requires the effect of viscosity to be considered in the equations which complicates the solution.In this paper,it is shown that the major contribution of viscosity and the horizontal component of actuation is concentrated in a narrow region alongside the actuation boundary.Since the inviscid equations are considerably easier to solve,a division into the viscous and inviscid domains would alleviate the computational load significantly.The particles'traces calculated by this approximation are excellently alongside their counterparts from the completely viscous model.It is also shown that the optimum thickness for the viscous strip is about 9-fold the acoustic boundary layer thickness for various flow patterns and amplitudes of actuation.

Significance of Live Streaming in Shaping Business: A Critical Review and Analytical Study

With the rise of live streaming on social media, platforms like Facebook, Instagram, and YouTube have become powerful business tools. They enable users to share live videos, fostering direct connections between businesses and their customers. This critical literature review paper explores the impact of live streaming on businesses, focusing on its role in attracting and satisfying consumers by promoting products tailored to their needs and wants. It emphasizes live streaming’s crucial role in engaging customers, a key to business growth. The study also provides viable strategies for businesses to leverage live streaming for growth and customer engagement, underscoring its importance in the business landscape.

Research on Current Situation and Legal Regulation of Cosmetics Live Streaming E-commerce in China

Analyze the compatibility between cosmetics and live streaming e-commerce from its own nature,marketing means and supply chain characteristics.According to the prominent problems,sort out the relationship between all parties in the cosmetics live e-commerce industry chain.Combined with the latest regulatory policies of live streaming e-commerce and cosmetics,the responsibilities of different subjects in cosmetics live streaming e-commerce are summarized,and relevant suggestions and countermeasures are put forward for the standardization and development of live streaming e-commerce.Cosmetics brand owners are the first responsible persons for product quality.Anchors,as a mixed identity between intermediary,advertising spokesperson and operator,should bear stricter joint and several liability when recommending products related to consumers’health.If anchors fail to clearly identify themselves in the recommendation process,thus causing consumers to mistake them for the operator of the cosmetics,they should assume the obligations of the operator.

基于Spark Streaming的车辆电子围栏技术实现与应用

文章提出一种基于Spark Streaming实时数据流处理框架,使用Kafka作为车辆轨迹数据的消息队列服务,结合拓扑关系判断算法射线法的车辆电子围栏技术。应用表明,该技术能够处理高吞吐率、强实时性的车辆动态数据,满足车辆动态精细化监管需求。

A Study on the Factors Influencing Consumer Purchase Decision Under the Live-Streaming Sales Model

In recent years,with the rapid development and popularization of Internet information technology,many new media platforms have risen rapidly,and major e-commerce companies have begun to explore the mode of livestreaming.Especially during the COVID-19 pandemic,due to the lockdown,live-streaming has become an important means of economic development in many places.Owing to its remarkable characteristics of timeliness,entertainment,and interactivity,it has become the latest and trendiest sales mode of e-commerce channels,reflecting huge economic potential and commercial value.This article analyzes two models and their characteristics of live-streaming sales from a practical perspective.Based on this,it outlines consumer purchasing decisions and the factors that affect consumer purchasing decisions under the live-streaming sales model.Finally,it discusses targeted suggestions for using the live-streaming sales model to expand the consumer market,hoping to promote the healthy and steady development of the live-streaming sales industry.

Machine Learning Based Classifiers for QoE Prediction Framework in Video Streaming over 5G Wireless Networks

Recently,the combination of video services and 5G networks have been gaining attention in the wireless communication realm.With the brisk advancement in 5G network usage and the massive popularity of threedimensional video streaming,the quality of experience(QoE)of video in 5G systems has been receiving overwhelming significance from both customers and service provider ends.Therefore,effectively categorizing QoE-aware video streaming is imperative for achieving greater client satisfaction.This work makes the following contribution:First,a simulation platform based on NS-3 is introduced to analyze and improve the performance of video services.The simulation is formulated to offer real-time measurements,saving the expensive expenses associated with real-world equipment.Second,A valuable framework for QoE-aware video streaming categorization is introduced in 5G networks based on machine learning(ML)by incorporating the hyperparameter tuning(HPT)principle.It implements an enhanced hyperparameter tuning(EHPT)ensemble and decision tree(DT)classifier for video streaming categorization.The performance of the ML approach is assessed by considering precision,accuracy,recall,and computation time metrics for manifesting the superiority of these classifiers regarding video streaming categorization.This paper demonstrates that our ML classifiers achieve QoE prediction accuracy of 92.59%for(EHPT)ensemble and 87.037%for decision tree(DT)classifiers.

基于Spark Streaming的气象自动站实时流处理与存储系统

在当前大数据技术蓬勃发展的时代,人们对气象数据的实时处理、数据质量、数据存储及大规模查询等要求也越来越高。针对现有气象自动站数据业务落地环节多,任务处理耦合紧但系统部署分散等问题,文中基于Spark Streaming的流式计算框架,研究使用Flume解析收集自动站原始数据,在Spark Streaming中设计融入自动站数据质控算法,最终通过对分布式数据库存储的表设计,使气象自动站数据具备高效率、高质量、高可靠的应用服务能力。性能测试结果表明,基于Spark Streaming的气象自动站数据实时流处理与存储系统,数据从文件采集、解码、流处理至入库的全流程能够在秒级完成,TB级数据查询响应为毫秒级,加权查询为秒级,完全满足自动站数据业务应用需求,从而为进一步提高气象自动站数据质量与服务水平提供基础支撑。

Distribution pattern of acoustic and streaming field during multi-source ultrasonic melt treatment process

The ultrasonic melt treatment(UMT)is widely used in the fields of casting and metallurgy.However,there are certain drawbacks associated with the conventional process of single-source ultrasonic(SSU)treatment,such as the fast attenuation of energy and limited range of effectiveness.In this study,the propagation models of SSU and four-source ultrasonic(FSU)in Al melt were respectively established,and the distribution patterns of acoustic and streaming field during the ultrasonic treatment process were investigated by numerical simulation and physical experiments.The simulated results show that the effective cavitation zone is mainly located in a small spherical region surrounding the end of ultrasonic horn during the SSU treatment process.When the FSU is applied,the effective cavitation zone is obviously expanded in the melt.It increases at first and then decreases with increasing the vibration-source spacing(Lv)from 30 mm to 100 mm.Especially,when the Lv is 80 mm,the area of effective cavitation zone reaches the largest,indicating the best effect of cavitation.Moreover,the acoustic streaming level and flow pattern in the melt also change with the increase of Lv.When the Lv is 80 mm,both the average flow rate and maximum flow rate of the melt reach the highest,and the flow structure is more stable and uniform,with the typical morphological characteristics of angular vortex,thus significantly expanding the range of acoustic streaming.The accuracy of the simulation results was verified by physical experiments of glycerol aqueous solution and tracer particles.

Adaptive Learning Video Streaming with QoE in Multi-Home Heterogeneous Networks

In recent years,real-time video streaming has grown in popularity.The growing popularity of the Internet of Things(IoT)and other wireless heterogeneous networks mandates that network resources be carefully apportioned among versatile users in order to achieve the best Quality of Experience(QoE)and performance objectives.Most researchers focused on Forward Error Correction(FEC)techniques when attempting to strike a balance between QoE and performance.However,as network capacity increases,the performance degrades,impacting the live visual experience.Recently,Deep Learning(DL)algorithms have been successfully integrated with FEC to stream videos across multiple heterogeneous networks.But these algorithms need to be changed to make the experience better without sacrificing packet loss and delay time.To address the previous challenge,this paper proposes a novel intelligent algorithm that streams video in multi-home heterogeneous networks based on network-centric characteristics.The proposed framework contains modules such as Intelligent Content Extraction Module(ICEM),Channel Status Monitor(CSM),and Adaptive FEC(AFEC).This framework adopts the Cognitive Learning-based Scheduling(CLS)Module,which works on the deep Reinforced Gated Recurrent Networks(RGRN)principle and embeds them along with the FEC to achieve better performances.The complete framework was developed using the Objective Modular Network Testbed in C++(OMNET++),Internet networking(INET),and Python 3.10,with Keras as the front end and Tensorflow 2.10 as the back end.With extensive experimentation,the proposed model outperforms the other existing intelligentmodels in terms of improving the QoE,minimizing the End-to-End Delay(EED),and maintaining the highest accuracy(98%)and a lower Root Mean Square Error(RMSE)value of 0.001.

A Novel Outlier Detection with Feature Selection Enabled Streaming Data Classification

Due to the advancements in information technologies,massive quantity of data is being produced by social media,smartphones,and sensor devices.The investigation of data stream by the use of machine learning(ML)approaches to address regression,prediction,and classification problems have received consid-erable interest.At the same time,the detection of anomalies or outliers and feature selection(FS)processes becomes important.This study develops an outlier detec-tion with feature selection technique for streaming data classification,named ODFST-SDC technique.Initially,streaming data is pre-processed in two ways namely categorical encoding and null value removal.In addition,Local Correla-tion Integral(LOCI)is used which is significant in the detection and removal of outliers.Besides,red deer algorithm(RDA)based FS approach is employed to derive an optimal subset of features.Finally,kernel extreme learning machine(KELM)classifier is used for streaming data classification.The design of LOCI based outlier detection and RDA based FS shows the novelty of the work.In order to assess the classification outcomes of the ODFST-SDC technique,a series of simulations were performed using three benchmark datasets.The experimental results reported the promising outcomes of the ODFST-SDC technique over the recent approaches.

基于效果定价模式的网络直播广告定价决策研究

网络直播广告作为一种新型营销方式快速发展,优化直播广告运营主体努力水平及定价策略是一项值得深入研究的课题。本文基于广告投放效果的两种定价模式,构建了包含两个广告商和一个主播的网络直播广告定价决策模型,探索广告商与主播的最优努力水平选择及广告定价策略。研究发现:CPW(cost per watch)定价模式下,广告商承担了消费者是否购买的不确定性风险,当消费者敏感性系数偏低时,广告商会提交较低的出价,且B/D两类广告商赢得竞拍的概率相等;对比CPW模式,在CPA(cost per action)定价模式下广告商的努力水平更低,且CPA定价模式中B型(品牌型)广告商赢得竞拍的概率更大,但赢得竞拍的广告商边际利润往往较低;与广告商相反,主播在CPA定价模式下的收益大于CPW,且随消费者敏感性系数的增加,两种定价模式下的收益差逐渐增大;CPW定价模式下预期观看直播的用户量和购买率均高于CPA,网络直播市场倾向于从CPW广告定价合同中获得较大收益。

面向不同类型概念漂移的两阶段自适应集成学习方法

大数据时代,流数据大量涌现.概念漂移作为流数据挖掘中最典型且困难的问题,受到了越来越广泛的关注.集成学习是处理流数据中概念漂移的常用方法,然而在漂移发生后,学习模型往往无法对流数据的分布变化做出及时响应,且不能有效处理不同类型概念漂移,导致模型泛化性能下降.针对这个问题,提出一种面向不同类型概念漂移的两阶段自适应集成学习方法(two-stage adaptive ensemble learning method for different types of concept drift,TAEL).该方法首先通过检测漂移跨度来判断概念漂移类型,然后根据不同漂移类型,提出“过滤-扩充”两阶段样本处理机制动态选择合适的样本处理策略.具体地,在过滤阶段,针对不同漂移类型,创建不同的非关键样本过滤器,提取历史样本块中的关键样本,使历史数据分布更接近最新数据分布,提高基学习器有效性;在扩充阶段,提出一种分块优先抽样方法,针对不同漂移类型设置合适的抽取规模,并根据历史关键样本所属类别在当前样本块上的规模占比设置抽样优先级,再由抽样优先级确定抽样概率,依据抽样概率从历史关键样本块中抽取关键样本子集扩充当前样本块,缓解样本扩充后的类别不平衡现象,解决当前基学习器欠拟合问题的同时增强其稳定性.实验结果表明,所提方法能够对不同类型的概念漂移做出及时响应,加快漂移发生后在线集成模型的收敛速度,提高模型的整体泛化性能.

基于数据流的K-S变化检测的动态多目标规划算法

为了更加准确地判断环境是否发生变化并快速追踪动态多目标规划问题(dynamicmulti-objectiveoptimization problem,DMOP)当前时刻的Pareto前沿,提出了一种基于数据流的Kolmogorov-Smirnov(K-S)变化检测的动态多目标规划(DSK-SDMOP)算法。该算法以NSGA-Ⅱ为基础,通过数据流建立2个时刻的检验窗口,再利用K-S检验基于数据流的Pareto最优前沿是否发生变化,检测2个窗口的数据是否服从同一分布来判断环境是否发生变化,并就环境变化的剧烈程度实行相应的应答机制,以提高对环境的适应程度。利用基于数据流的K-S检测方法,对环境变化不会过于敏感,而且不用提前假设对应目标值的分布,易于操作。通过5个动态多目标规划标准测试函数对该算法进行测试,并和现有的2种算法进行对比分析,结果表明该算法处理动态多目标规划问题具有良好的性能。

教育建筑的建筑教育——以本科四年级学校设计教学实践为例

从东南大学建筑学院本科四年级教育建筑设计系列课程的教学实践出发,对教育制度和空间模式相互关系展开深入研究。教育建筑设计系列课程通过立足社会问题的题目设置,职业建筑师的介入以及交叉学科的设计研究来指导教学实践,研究与总结其成果,对于发展更具有创新意识及职业性的建筑教育有重要的参考和借鉴意义。

多源流理论视角下美国职业生涯教育立法政策的变迁

多源流理论为解释政策过程和分析政策变迁提供了一种视角和框架。美国职业生涯教育立法政策经历了三个时期:一是以片段化的引导性立法政策支持职业指导和职业生涯发展的探索期;二是以特色化的规制性立法政策推进职业生涯教育实践的确立期;三是以系统化的策略性立法政策深化职业生涯教育改革的深化期。运用多源流理论分析这一变迁历程会发现:美国时代变动引发的失业危机与传统教育效能单一之间的冲突构成了“问题识别”;职业生涯教育思想在政策共同体中的演化促成了“政策阐明”;公众舆论与政党执政理念对职业生涯教育的影响产生了“政治活动”;马兰德的“过程软化”与打开“政策之窗”则最终促成了“政策之窗与溪流的结合”。回顾美国职业生涯教育立法政策的颁布与调整,有两点值得思考和借鉴:一是立法政策在发展趋势、人群面向、目标定位和功能发挥上有明显变化;二是立法政策规范和保障是影响美国职业生涯教育实践的重要推动因素,而职业生涯教育立法政策的颁布又是特定时期多种因素长期共同作用促成的结果。

二重情境:数字视听文化中的身份构建与认同疏离

数字视听媒介及其文化的发展创造了新的情境,推动用户产生新的媒介实践,但也带来了新的问题和现象,即情境生成逻辑向流媒体用户让渡的转变、情境功能作为交往工具的偏移,以及情境秩序在去公共性过程中权力再结构化的取向。其中隐含着情境之于用户身份构建的二重性:一方面,流媒体用户依据情境构建身份,身份的构建过程进一步激发了情境的创造;另一方面,这些被构建的身份呈现出去情境化的流动趋向,造成了身份与认同的疏离。对此,文章在厘清数字视听文化新情境特征的基础上,阐释流媒体用户基于身份的情境互动过程,以及身份认同疏离的成因及危机。更进一步,文章尝试将新媒介—新情境—新行为的线性模式延展为更具解释力的用户主导循环模式,并提出通过共识的凝聚、公共性的重拾和共同体的重建,推动数字视听文化的良序发展。

基于振荡射流的撞击流反应器流动及混合特性研究

为了研究振荡射流撞击流反应器内流场特性,利用SST k-ω湍流模型,建立振荡射流撞击流反应器数学模型且与试验进行对比。通过分析速度场、湍流动能分布、压力场和湍流黏度分布规律来研究撞击流反应器内部流动特性,揭示混合性能变化规律。结果表明:在轴向上,速度、湍动能和压力呈多峰分布趋势且随着入口速度的上升而不断增大。在径向上,振荡器混合室内部涡旋中心受喷嘴小尺度涡旋干扰不断变化,从而影响流体振荡,使流场内产生4种流型。该反应器在35 s后混合均匀,混合强度达到0.95以上。研究结果丰富了流体振荡特性理论,为新型反应装置的开发提供理论参考。

虚拟直播电商的销售业绩影响因素与发展对策研究——基于信息源理论的实证分析

随着人工智能技术的发展,由虚拟主播与虚拟场景结合而成的虚拟直播成为电商发展新方向。基于直播电商“流量+转化”的商业本质,在信息源理论框架下构建虚拟直播电商的销售业绩模型,以淘宝直播平台300家虚拟直播间为样本进行实证分析。结果发现:虚拟主播的可信性、专业性和吸引力对销售业绩具有显著的正向影响,作为直播电商主要特征的互动性反而对销售业绩具有显著的负向影响,商品类型对虚拟主播互动性具有调节作用。根据实证分析结果,建议强化虚拟主播特性以提高流量和转化率,提升技术应用水平以填补互动性短板,出台培育政策以降低虚拟主播应用成本。

非平衡数据流在线主动学习方法

数据流分类是数据流挖掘领域一项重要研究任务,目标是从不断变化的海量数据中捕获变化的类结构.目前,几乎没有框架可以同时处理数据流中常见的多类非平衡、概念漂移、异常点和标记样本成本高昂问题.基于此,提出一种非平衡数据流在线主动学习方法(Online active learning method for imbalanced data stream,OALM-IDS).AdaBoost是一种将多个弱分类器经过迭代生成强分类器的集成分类方法,AdaBoost.M2引入了弱分类器的置信度,此类方法常用于静态数据.定义了基于非平衡比率和自适应遗忘因子的训练样本重要性度量,从而使AdaBoost.M2方法适用于非平衡数据流,提升了非平衡数据流集成分类器的性能.提出了边际阈值矩阵的自适应调整方法,优化了标签请求策略.将概念漂移程度融入模型构建过程中,定义了基于概念漂移指数的自适应遗忘因子,实现了漂移后的模型重构.在6个人工数据流和4个真实数据流上的对比实验表明,提出的非平衡数据流在线主动学习方法的分类性能优于其他5种非平衡数据流学习方法.