Toggle navigation
Home
About
About Journal
Historical Evolution
Indexed In
Awards
Reference Index
Editorial Board
Journal Online
Archive
Project Articles
Most Download Articles
Most Read Articles
Instruction
Contribution Column
Author Guidelines
Template
FAQ
Copyright Agreement
Expenses
Academic Integrity
Contact
Contact Us
Location Map
Subscription
Advertisement
中文
Journals
Publication Years
Keywords
Search within results
(((LONG Xia[Author]) AND 1[Journal]) AND year[Order])
AND
OR
NOT
Title
Author
Institution
Keyword
Abstract
PACS
DOI
Please wait a minute...
For Selected:
Download Citations
EndNote
Ris
BibTeX
Toggle Thumbnails
Select
Graph trend filtering guided noise tolerant multi-label learning model
LIN Tengtao, ZHA Siming, CHEN Lei, LONG Xianzhong
Journal of Computer Applications 2021, 41 (
1
): 8-14. DOI:
10.11772/j.issn.1001-9081.2020060971
Abstract
(
414
)
PDF
(972KB)(
547
)
Knowledge map
Save
Focusing on the problem that the feature noise and label noise often appear simultaneously in multi-label learning, a Graph trend filtering guided Noise Tolerant Multi-label Learning (GNTML) model was proposed. In the proposed model, the feature noise and label noise were tolerated at the same time by group sparsity constraint bridged with label enrichment. The key of the model was the learning of the label enhancement matrix. In order to learn a reasonable label enhancement matrix in the mixed noise environment, the following steps were carried out. Firstly, the Graph Trend Filtering (GTF) mechanism was introduced to tolerate the inconsistency between the noisy example features and labels, so as to reduce the influence of the feature noise on the learning of the enhancement matrix. Then, the group sparsity constrained label fidelity penalty was introduced to reduce the impact of label noise on the label enhancement matrix learning. At the same time, the sparsity constraint of label correlation matrix was introduced to characterize the local correlation between the labels, so that the example labels were able to propagate better between similar examples. Finally, experiments were conducted on seven real multi-label datasets with five different evaluation criteria. Experimental results show that the proposed model achieves the optimal value or suboptimal value in 66.67% cases, it is better than other five multi-label learning algorithms, and can effectively improve the robustness of multi-label learning.
Reference
|
Related Articles
|
Metrics
Select
Fish swarm algorithm optimized by PSO applied in maximum power point tracking of photovoltaic power system
DUAN Qi-chang TANG Ruo-li LONG Xia
Journal of Computer Applications 2012, 32 (
12
): 3299-3302. DOI:
10.3724/SP.J.1087.2012.03299
Abstract
(
817
)
PDF
(563KB)(
587
)
Knowledge map
Save
Introducing the velocity inertia, memory capacity of each individual and learning or communicating capacity of Particle Swarm Optimization (PSO) into the Artificial Fish-Swarm Algorithm (AFSA), a new algorithm called the “Fish-Swarm Algorithm optimized by PSO(PSO-FSA)” was put forward. In this new algorithm, the swimming of each fish has velocity inertia, and the PSO-FSA has totally five kinds of behavior pattern as follows: swarming, following, remembering, communicating and searching. The simulation analysis shows that PSO-FSA has more stable and higher performance in convergence speed and searching precision than PSO and AFSA. Finally, the PSO-FSA was applied to the maximum power point tracking of photovoltaic power generation system under partially shaded condition, and the experimental results show that PSO-FSA can find the maximum power point under partially shaded insolation conditions quickly and precisely.
Related Articles
|
Metrics
Select
Extraction technology of blog comments based on functional semantic units
FAN Chun-long XIA Jia XIAO Xin LV Hong-wei XU Lei
Journal of Computer Applications 2011, 31 (
09
): 2417-2420. DOI:
10.3724/SP.J.1087.2011.02417
Abstract
(
1269
)
PDF
(813KB)(
507
)
Knowledge map
Save
Blog is an important kind of network information resources, and the extraction of its comments is the basic work of public opinion analysis researches and of such work. The current mainstream blog comments extraction algorithms were summarized, and the application of page structure in information extraction was described. Using the characteristics of indicating phrases such as the "Home" when people understand Web pages, technology of extracting comments information was proposed by utilizing functional semantic units that they have clear semantics and functional indication. Many technologies involved in the extraction process were detailed such as page structure linearization, functional semantic units recognition, text distinguishment and comments extraction algorithm. Finally, the experimental results show that this technology can achieve better results in extraction of blog body and comments.
Related Articles
|
Metrics