Dynamic multi%2Dcue tracking with detection responses association.

Automatic interesting object extraction from images using complementary saliency maps.

Saliency detection based on 2D log%2Dgabor wavelets and center bias.

ESUR A system for Events detection in SURveillance video.

Video retargeting with multi%2Dscale trajectory optimization.

Vlogging A survey of videoblogging technology on the web.

Social Multimedia Computing.

Probabilistic Multi%2DTask Learning for Visual Saliency Estimation in Video.

Salient object extraction for user%2Dtargeted video content association.

A dataset and evaluation methodology for visual saliency in video.

Multiple kernel active learning for image classification.

A New Multiple Kernel Approach for Visual Concept Learning.

DCT%2DBased Videoprinting on Saliency%2DConsistent Regions for Detecting Video Copies with Text Insertion.



Multi%2Dpolarity text segmentation using graph theory.

Towards multi%2Dgranularity multi%2Dfacet e%2Dbook retrieval.

Diversifying the image retrieval results.

Robust Collective Classification with Contextual Dependency Network Models.

Semantic Scoring Based on Small%2DWorld Phenomenon for Feature Selection in Text Mining.

Keyphrase Extraction Using Semantic Networks Structure Analysis.
Keyphrases play a key role in text indexing, summarization, and categorization. However, most of the existing keyphrase extraction approaches require human-labeled training sets. In this paper, we propose an automatic keyphrase extraction algorithm using two novel feature weights, which can be used in both supervised and unsupervised tasks. This algorithm treats each document as a semantic network that holds both syntactic and statistical information. Structural dynamics of these networks can easily identify key nodes, which can be used to extract keyphrases unsupervisedly. Experiments demonstrate the proposed keyphrase extraction algorithm averagely improves 50 % in effectiveness and 30 % in efficiency in unsupervised tasks and performs comparatively with supervised extractors. Moreover, by applying this algorithm to supervised task, we develop a more accurate classifier. In this classifier, we assemble several syntactic and statistical features. Experiments show that the overall precision of supervised extraction can be up to 80%. 1.
Improving the Image Retrieval Results Via Topic Coverage Graph.

Context%2Dbased statistical relational learning.

Learning Contextual Dependency Network Models for Link%2DBased Classification.
Links among objects contain rich semantics that can be very helpful in classifying the objects. However, many irrelevant links can be found in real-world link data such as Web pages. Often, these noisy and irrelevant links do not provide useful and predictive information for categorization. It is thus important to automatically identify which links are most relevant for categorization. In this paper, we present a contextual dependency network (CDN) model for classifying linked objects in the presence of noisy and irrelevant links. The CDN model makes use of a dependency function that characterizes the contextual dependencies among linked objects. In this way, CDNs can differentiate the impacts of the related objects on the classification and consequently reduce the effect of irrelevant links on the classification. We show how to learn the CDN model effectively and how to use the Gibbs inference framework over the learned model for collective classification of multiple linked objects. The experiments show that the CDN model demonstrates relatively high robustness on data sets containing irrelevant links.
Context%2DBased Classification for Link Data.

Two%2Dphase Web site classification based on Hidden Markov Tree models.

Two%2DPhase Web Site Classification Based on Hidden Markov Tree Models.

Quantitatively evaluating the influence of online social interactions in the community%2Dassisted digital library.

