Zobrazeno 1 - 10
of 20
pro vyhledávání: '"Shifeng Weng"'
Publikováno v:
Neurocomputing. 179:26-36
Nystrom method has been widely used to improve the computational efficiency of batch kernel learning. The key idea of Nystrom method is to randomly sample M support vectors from the collection of T training instances, and learn a kernel classifier in
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::5a247bb8302859fad031b382a0132032
https://doi.org/10.1016/j.neucom.2015.10.002
https://doi.org/10.1016/j.neucom.2015.10.002
Publikováno v:
Neurocomputing. 149:100-105
Support Vector Data Description (SVDD) is an important algorithm for data description problem. SVDD uses only positive examples to learn a predictor whether an example is positive or negative. When a fraction of negative examples are available, the p
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::463aee6a7adfc4a008e6388d382e5a7a
https://doi.org/10.1016/j.neucom.2014.03.072
https://doi.org/10.1016/j.neucom.2014.03.072
Publikováno v:
Neural Computing and Applications. 26:957-968
I-MLOF algorithm is an extension of local outlier factor (LOF) algorithm in multiple instance (MI) setting. The task of I-MLOF is to identify MI outlier. However, I-MLOF algorithm works in batch mode, where all samples must be provided for once. In s
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3d1926857a3f2d54c2b4818befe57216
https://doi.org/10.1007/s00521-014-1750-6
https://doi.org/10.1007/s00521-014-1750-6
Publikováno v:
International Journal of Hybrid Information Technology. 7:309-320
Users’ similarity mining in mobile e-commerce systems is an important field with wide applications, such as personalized recommendation and accurate advertising. Moving trajectories of e-commerce users contain much useful information, providing a v
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::1c1882234066cfac464a653e3d954092
https://doi.org/10.14257/ijhit.2014.7.4.26
https://doi.org/10.14257/ijhit.2014.7.4.26
Publikováno v:
ACM Transactions on Knowledge Discovery from Data. 8:1-19
We study the sample complexity of random Fourier features for online kernel learning—that is, the number of random Fourier features required to achieve good generalization performance. We show that when the loss function is strongly convex and smoo
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a3a65f673d6a7186710d15af6fb4a840
https://doi.org/10.1145/2611378
https://doi.org/10.1145/2611378
Publikováno v:
Medical & Biological Engineering & Computing. 43:410-412
The paper describes an application of a new, non-linear dimensionality reduction method, named Isomap, for mining the structural knowledge from high-dimensional medical data. The algorithm was evaluated on two publicly available medical datasets: the
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::dba1089825f4de7437151aa508fa08fd
https://doi.org/10.1007/bf02345820
https://doi.org/10.1007/bf02345820
Publikováno v:
Pattern Recognition. 38:599-601
In this paper, we investigated nonlinear dimensionality reduction (NLDR) for supervised learning and introduced a novel algorithm named supervised isometric mapping (SIsomap) which was based on a combination of two well-known methods: isomap and fuzz
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a6a8ca07832442410e3b48798df3355b
https://doi.org/10.1016/j.patcog.2004.08.014
https://doi.org/10.1016/j.patcog.2004.08.014
Publikováno v:
ACPR
We propose a set of efficient processes for extracting all four elements of Chinese news web pages, namely news title, release date, news source and the main text. Our approach is based on a deep analysis of content and structure features of current
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::21bb338b2788e72337b31864e90972c6
https://doi.org/10.1109/acpr.2013.52
https://doi.org/10.1109/acpr.2013.52
Publikováno v:
IJCNN
In real world applications, we often have to deal with some high-dimensional, sparse and noisy data. In this paper, we aim to handle this kind of complex data by a Robust Non-negative Matrix Factorization via joint Sparse and Graph regularization mod
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::45d002d539ee11bbe587cc37f4c36f5f
https://doi.org/10.1109/ijcnn.2013.6706859
https://doi.org/10.1109/ijcnn.2013.6706859
Publikováno v:
ChinaSIP
Modern internet technologies make people easily access millions of songs. However it also forms a huge barrier between customers and those songs people truely want, due to the difficulty to explore this large collections. In this paper, we propose a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::9f859f17d3e81497602ed7bbcf3f404c
https://doi.org/10.1109/chinasip.2013.6625322
https://doi.org/10.1109/chinasip.2013.6625322