融合深度情感分析和评分矩阵的推荐模型

目前,大多数推荐系统都具有评分数据稀疏性的问题,它会限制模型的有效性。而用户对于某件商品撰写的评论中隐含了很多信息,对评论文本进行情感分析并提取关键的因素来用于模型的学习,可以有效地缓解数据稀疏问题,但仅使用评论数据而忽略了评分数据的主要因素会影响推荐精度。对此,为了进一步提高推荐精度,该文提出一个评论文本和评分矩阵交互(RTRM)的深度模型,该模型能够提取评论文本和评分矩阵的深层次特征,并结合它们进行评分预测;其次,通过使用预训练的Electra模型得到每条评论的隐表达,并结合深度情感分析及注意力机制实现从上下文语义层面对评论文本的分析,解决了短文本的语义难以分析的问题;同时,在融合层模块中,用户(物品)评论和评分矩阵进行交互,最终预测出用户对商品的评分;最后,在6组数据集上,采用均方误差(MSE)进行性能对比实验,实验结果表明该文模型性能优于其他系统,且平均预测误差最大降低了12.821%,该模型适用于向用户推荐精确的物品。     

基于Bert预训练模型的虚假新闻文本检测

虚假新闻剥夺了人们获取真相的权利,也给社会和国家带来了许多危害。文中以虚假新闻文本为例,分析和验证了多种预训练语言模型在虚假新闻文本分类上的分类效果。经实验证明,相较于其他语言模型,Bert预训练语言模型取得了较好的结果,预测准确率为86.97%,召回率为88.12%,F1值为87.54%。     

A path planning model based on nested regular hexagons using weighted centroid localization

In many wireless sensor network (WSN) applications, the location of a sensor node is crucial for determining where the event or situation of interest occurred. Therefore, localization is one of the critical challenges in WSNs. Mobile anchor node assisted localization (MANAL) is one of the promising solutions for the localization of statically deployed sensors. The main problem in MANAL localization is that the path planning of the mobile anchor (MA) node should be done so that the localization error in the network will be minimal and that all unknown nodes in the network are covered. This paper proposes a new path planning approach called nested hexagons curves (NHexCurves) for MANAL. NHexCurves guarantees that it will receive messages from at least three non-collinear anchors to locate all unknown nodes in the network. The proposed model has compared six different path planning schemes in the literature using weighted centroid localization (WCL). In these comparisons, first of all, localization errors of the models are compared using some statistical concepts. Second, the variation of the localization error according to parameters such as resolution (R) and the standard deviation of noise (σ) is observed. Then, with similar approaches, the standard deviation of errors, localization ratio, scalability performances, and finally, path lengths of the models are examined. The simulation results show that the NHexCurves static path planning model proposed in this study stands out compared to other models with high localization error and localization ratio performance, especially at low resolutions, due to its path design. At the same time, the lowest error values according to σ are obtained with the proposed model among all models considered.     

基于改进YOLOv3的快速文本检测

针对深度学习文本检测算法存在运行速度慢、模型体积大等问题,提出了基于改进的YOLOv3(You Only Look Once v3)文本检测方法(mobile-text-YOLOv3).通过深度可分离卷积思想轻量化Darknet-53网络,在高层特征借助双线性插值和偏移层使卷积核具有可变感受野,较大地改善了模型的性能;...     

基于斜率密度聚类的相似文本标定

相似文本标定是抄袭检测的一个重要环节,现有标定方法大多采用直接对文本或指纹进行合并的方式,标定精度受干扰信息影响较大。针对这种局限性,分析了匹配指纹对的语义特征,提出基于斜率密度的相似文本聚类方法,将文本匹配合并问题转化成稠密样本点聚类问题,并在PAN公用语料库上对该方法进行了测试,得到的主要指标优于PAN10前3名。目前已将该方法用于华南理工大学特色专业教学平台的作业查抄,取得了较好的效果。     

基于改进单锚节点的无线传感器网络节点定位算法

节点定位是无线传感器关键技术之一,针对固定多锚节点方法定位精度低的缺陷,为了提高无线传感器的定位精度,提出了一种基于改进单锚节点的无线传感器网络节点定位算法(SFOA-SVM)。首先采用单移动锚节点在无线传感器网络中移动,构建无线传感器定位模型的学习样本,然后采用SVM构建节点定位模型,并采用渔夫捕鱼算法模拟渔夫捕鱼行为找到最优SVM参数,最后采用仿真实验测试节点的定位性能。结果表明,相对于其它定位算法,SFOA-SVM提高了无线传感器节点的定位精度,具有一定的实际应用价值。     

基于条件随机场的复杂背景图像文字分割

图像中的文字提取是处理文字信息的关键步骤。提出一种基于条件随机场（CRFs）的分割方法。针对背案复杂的情况,精心选择特征,设计分类器,计算在给定观测数据情况下,像素标签的条件概率分布。这样,有效地避免了计算联合分布所带来的问题（如马尔可夫随机场）。与传统的分类器相比,该方法更加注重像素以及标签之间互相影响程度以及合理性。试验结果表明,与其他方法相比,CRFs的文字分割结果有明显的提高。     

DMAP: integrated mobility and service management in mobile IPv6 systems

Mobile IPv6 (MIPv6) is a work in progress IETF standard for enabling mobility in IPv6 networks and is expected to have wide deployment. We investigate an integrated mobility and service management scheme based on MIPv6 with the goal to minimize the overall network signaling cost in MIPv6 systems for serving mobility and service management related operations. Our design extends IETF work-in-progress Hierarchical Mobile IPv6 (HMIPv6) with the notion of dynamic mobility anchor points (DMAPs) for each mobile node (MN) instead of static ones for all MNs. These DMAPs are access routers chosen by individual MNs to act as a regional router to reduce the signaling overhead for intra-regional movements. The DMAP domain size, i.e., the number of subnets covered by a DMAP, is based on the MN’s mobility and service characteristics. Under our DMAP protocol, a MN interacts with its home agent and application servers as in the MIPv6 protocol, but optimally determines when and where to launch a DMAP to minimize the network cost in serving the user’s mobility and service management operations. We demonstrate that our DMAP protocol for integrated mobility and service management yields significantly improved performance over basic MIPv6 and HMIPv6.     

面向离散文本舆情分析的分聚类方案

离散文本已经成为一种占据重要地位的舆情信息表现形式,根据离散文本的特点,提出基于特征概念网的离散文本舆情信息的分聚类框架,在此基础上给出分聚类方案。在聚类算法中,运用了遗传算法的全局并行搜索能力、k—means的高效局部聚类能力和小生境的保持种群多样性抑制漂移能力;在分类算法中,先将训练文本库进行类内聚类成子类,对子类构建特征概念网以生成替代该子类的文本,再用KNN算法进行分类。最后结合舆情分析进一步提出了可用的改进方案。     

应用三维弹簧系统模型的锚节点部署优化

为使随机部署的三维无线传感器网络中锚节点的分布更加合理,提高未知节点定位精度,针对锚节点部署进行优化。通过构建弹簧系统模型,将锚节点抽象为通过弹簧相连接的点,使部分锚节点在合力作用下进行伸缩运动,达到提高网络性能的目的。当锚节点部署优化完成后,应用近似三角形内点测试( APIT)和DV-HOP( Distance Vector-hop)算法测试优化前后的节点定位精度。仿真结果表明,三维空间下的锚节点经过弹簧系统模型的部署优化后,锚节点网络覆盖率和定位覆盖率均得到了提高,网络平均连通度有所提升,且定位精度显著提高。