Results 11 to 20 of about 13,106 (168)
智能科学与技术学报, 2021 与中英文相比，藏文实体关系训练语料规模较小，传统有监督的学习方法难以获得较高的准确率。针对基于远程监督的实体关系抽取存在错误标记的问题，利用远程监督方法将知识库与文本对齐，构建藏文实体关系抽取的数据集，提出一个基于多级注意力融合机制的藏文实体关系抽取模型。在词级别引入自注意力机制来提取单词的内部特征，在句子级别引入注意力机制为每个实例分配权重，从而充分利用包含信息的句子，减少噪声实例的权重。同时引入联合评分函数，修正远程监督的错误标签，并将神经网络与支持向量机结合，实现藏文实体关系分类。实验结果表明 ...
王丽客, 孙媛, 刘思思
doaj
智能科学与技术学报, 2022 当前利用社交媒体数据预测抑郁症的研究忽视语言风格和情感随时间推移变化的特征，缺乏对情感状态和帖子元数据特征的研究，基于此，提出一种基于情感信息融合注意力机制的抑郁症识别模型。首先，在现有抑郁症识别研究的基础上，使用文本分类卷积神经网络按时间顺序提取用户每个时间段的帖子信息和情感信息，引入注意力机制为得到的特征矩阵分配不同的注意力权重，进而得到用户帖子信息特征和用户情感信息特征。其次，使用正则匹配提取文本的情感倾向信息，拼接注意力机制输出分配权重后的情感特征矩阵，以增强情感学习表示。然后 ...
陈妍, 罗雪琴, 梁伟, 谢永芳
doaj
Zhejiang Daxue xuebao. Lixue ban, 2023 提出了一种新颖的广义无监督函数映射学习的三维形状密集对应方法。首先，基于多层感知器（multilayer perceptron，MLP）和残差网络，直接学习深度点特征。其次，计算点云的近似测地线距离，并对其进行特征分解，建立特征嵌入空间，引入注意力机制，有效学习广义基函数表示。再次，结合点特征与广义基函数生成三维形状的深度特征表示。最后，建立无监督的函数映射网络框架，获取形状之间的密集对应表示。提出的三元正则优化机制，联合重构损失、特征损失和形状匹配的距离损失 ...
窦丰(DOU Feng) +6 more
doaj +1 more source
Research on Entity Relationship Extraction Model of Food Public Opinion Based on CNN-BLSTM(基于CNN-BLSTM的食品舆情实体关系抽取模型研究) [PDF]
Shipin kexue jishu xuebao, 2021 The extraction of food public opinion entity relationship is the key technology for constructing the map of food sensation knowledge, and it is also an important research topic in the field of information extraction.
WANG Qingbang(王庆棒) +5 more
doaj +1 more source
Aerial Violence Recognition Based on Spatial-Temporal Graph Convolutional Networks and Attention Model [PDF]
, 2022 The violence in public areas occurs frequently and video surveillance is of great significance for maintaining public safety.Compared with fixed cameras,unmanned aerial vehicles (UAVs) have surveillance mobility.However,in aerial images,the rapid ...
SHAO Yan-hua, LI Wen-feng, ZHANG Xiao-qiang, CHU Hong-yu, RAO Yun-bo, CHEN Lu
core +1 more source
Class Attention to Regions of Lesion for Imbalanced Medical Image Recognition
, 2023 Automated medical image classification is the key component in intelligent diagnosis systems. However, most medical image datasets contain plenty of samples of common diseases and just a handful of rare ones, leading to major class imbalances. Currently,
Cai, Jiabin +4 more
core
YOLOv3-A: a traffic sign detection network based on attention mechanism [PDF]
, 2021 To solve the problem that the existing YOLOv3 algorithm had more false detections and missed detections for traffic sign detection task with small target problems and complex background, based on the YOLOv3, a channel attention method for target ...
Fan GUO +3 more
core +1 more source
智能科学与技术学报, 2021 针对多光谱行人检测系统存在特征融合质量低、模型超参数多且锚框匹配算法复杂等问题，提出了一种基于差分特征注意力机制的无锚框多光谱行人检测算法。该算法首先采用差分特征感知融合方法挖掘多模态特征间的互补信息来优化通道特征；然后利用具有高效无锚框机制的CenterNet检测框架大大降低了模型计算复杂度，从而提升检测速度；最后引入差分特征注意力机制，改善特征融合质量，进一步提升检测精度。在KAIST、CVC14和FLIR这3个公开数据集上的实验结果表明，提出的算法和其他先进方法相比，能够同时有效提升检测精度和速度，
沈继锋 +4 more
doaj
, 2022 Video understanding is an important task in short video business platforms and it has a wide application in video recommendation and classification. Most of the existing video understanding works only focus on the information that appeared within the ...
Deng, Jiaxin +9 more
core
Hybrid Recommender System Based on Attention Mechanisms and Gating Network [PDF]
, 2022 Combining user reviews with user ratings to improve the performance of recommender system is the current mainstream research direction of recommender system.However,when user review data is sparse,the performance of most existing recommender systems will
GUO Liang, YANG Xing-yao, YU Jiong, HAN Chen, HUANG Zhong-hao
core +1 more source