Digital media Computing & Design Lab初建于1998年。团队围绕跨媒体计算、人工智能、大数据理论与应用三大方向,以合作承担国家重大(重点)项目为契机,历经将近20年建设,形成了优秀研究群体。团队核心成员以中青年人才为主,年富力强,开拓创新,活跃在教学科研第一线,拥有深厚理论功底和丰富科研经验,具有以国际学科前沿、国家需求和实际应用为导向的社会责任感,也具有充沛时间和精力完成好本团队科研任务。团队研究具有明显跨学科和交叉特色,为研究内容顺利实施奠定了学科交叉人才基础。
IEEE Transactions on Neural Networks and Learning Systems, no. 3 (2018): 718-730
It is observed that distinct words in a given document have either strong or weak ability in delivering facts (i.e., the objective sense) or expressing opinions (i.e., the subjective sense) depending on the topics they associate with. Motivated by the intuitive assumption that di...
Cited by12BibtexViews29Links
0
0
IEEE Transactions on Circuits and Systems for Video Technology, no. 1 (2018): 76-89
For a number of important problems, isolated semantic representations of individual syntactic words or visual objects do not suffice, but instead a compositional semantic representation is required; for example, a literal phrase or a set of spatially concurrent objects. In this p...
Cited by8BibtexViews15Links
0
0
TKDD, no. 1 (2018): 12:1-12:23
We find a “10-minute phenomenon” from the Twitter data, as shown in Figure 4(a), if two companies are mentioned by a user in her tweet(s) in 10 minutes, there is a likelihood of 44% that the two companies are competitors, which is 25 times higher than chance
Cited by5BibtexViews82Links
1
0
Frontiers of IT & EE, no. 1 (2017): 3-14
In this paper, we review recent emerging theoretical and technological advances of artificial intelligence (AI) in the big data settings. We conclude that integrating data-driven machine learning with human knowledge (common priors or implicit intuitions) can effectively lead to ...
Cited by47BibtexViews20Links
0
0
AAAI, pp.3532-3539, (2017)
We develop an asymmetric ranking network learning method with deep recurrent neural networks by integrating both answers’ relative quality rank to the given question and the answerers’ following relations in community-based question answering sites
Cited by38BibtexViews55Links
0
0
AAAI, pp.3119-3126, (2017)
This paper presents KGE-latent Dirichlet allocation, which combines topic model and knowledge graph embeddings, in particular latent Dirichlet allocation model and TransE
Cited by24BibtexViews24Links
0
0
ACM Trans. Inf. Syst., no. Issue-in-Progress (2017)
We model users’ social decisions on connecting and maintaining relationships conditioned on their demographic profiles in large-scale mobile communication networks
Cited by12BibtexViews42Links
0
0
IEEE Transactions on Circuits and Systems for Video Technology, no. 7 (2017): 1464-1477
In general, an image attribute is a human-nameable visual property that has a semantic connotation. Appropriate modeling of the intrinsic contextual correlations among attributes plays a fundamental role in attribute detection. In this paper, we consider image attribute detection...
Cited by12BibtexViews14Links
0
0
Neurocomputing, (2017): 58-70
Abstract In this paper, we present a novel document summarization mechanism called KeyphraseDS that can organize the scientific articles into multi-aspect and informative scientific survey by exploiting keyphrases. Keyphrases describe textu0027s salience and central focus, whic...
Cited by12BibtexViews10Links
0
0
IEEE Transactions on Knowledge and Data Engineering, no. 9 (2017): 1916-1927
We devise a Bregman distance function for the ranking model, and formulate the problem of learning Bregman distance to rank as a structural learning problem based on structural SVM
Cited by5BibtexViews7Links
0
0
computer vision and pattern recognition, (2016)
We proposed a new method, namely Hierarchical Recurrent Neural Encoder, to generate video representation with emphasis on temporal modeling
Cited by276BibtexViews13Links
0
0
IEEE Trans. Knowl. Data Eng., no. 9 (2016): 2522-2534
A social recommendation system has attracted a lot of attention recently in the research communities of information retrieval, machine learning, and data mining. Traditional social recommendation algorithms are often based on batch machine learning methods which suffer from sever...
Cited by63BibtexViews13Links
0
0
IJCAI, pp.3000-3006, (2016)
We propose the heterogenous Community-based question answering network that exploits both users’ relative quality rank to the given questions and their social relations for expert finding
Cited by61BibtexViews26Links
0
0
AAAI, pp.122-128, (2016)
We argue that the utilization of social information can mitigate the sparsity problem in Community-based question answering tasks
Cited by53BibtexViews43Links
0
0
IEEE Trans. Knowl. Data Eng., no. 3 (2016): 689-700
Feature selection is a challenging problem for high dimensional data processing, which arises in many real applications such as data mining, information retrieval, and pattern recognition. In this paper, we study the problem of unsupervised feature selection. The problem is chall...
Cited by53BibtexViews19Links
0
0
COLING, pp.641-651, (2016)
Most existing methods regard knowledge graph as a set of independent triples, and ignore the indirect dependency between subjects. To deal with this issue, we propose a novel method, graph aware knowledge embedding method, for learning the representation of a given knowledge grap...
Cited by49BibtexViews43Links
0
0
IEEE Trans. Image Processing, no. 12 (2016): 5892-5904
The proposed deep learning framework simultaneously performs deep feature learning for visual representation in conjunction with spatiotemporal context modeling, which largely enhances the scene understanding capability
Cited by36BibtexViews15Links
0
0
IJCAI, pp.3545-3551, (2016)
We propose and evaluate a unified approach for person Re-ID that jointly learns a deep convolutional neural network for representing the semantics-aware information of the image as well as a network for robust image matching based on correspondence structure learning
Cited by30BibtexViews9Links
0
0
AAAI, pp.65-71, (2016)
We find users with social roles of opinion leaders and structural hole spanners tend to be more influential than ordinary users in positive emotion contagion while be less influential in negative emotion contagion
Cited by26BibtexViews68Links
0
0
IEEE Transactions on Cybernetics, no. 7 (2016): 1498-1510
A key problem in level set tracking is to construct a discriminative speed function for effective contour evolution. In this paper, we propose a level set tracking method based on a discriminative speed function, which produces a superpixel-driven force for effective level set ev...
Cited by14BibtexViews3Links
0
0
Keywords
Feature ExtractionLearning (artificial Intelligence)Data MiningData ModelsDigital LibrarySemanticsSupport Vector MachinesVisual TrackingComputer ScienceGraph Theory
Authors
Yueting Zhuang
Paper 51
Xi Li
Paper 43
Fei Wu
Paper 37
Weiming Lu
Paper 28
Baogang Wei
Paper 27
Weiming Hu
Paper 17
Siliang Tang
Paper 16
Zhongfei (Mark) Zhang
Paper 15
Yin Zhang
Paper 13
Jie Tang
Paper 11