Y

Yang, Xiaoxia; Chen, Anmin; Ma, Yingfu; Gao, Yuan; Gao, Zhemin; Fu, Bingyi; Sun, Fuchun; Qiao, Jie; Li, Quan; Wan, Shuqian; Werner, Hermut; Chuang, Yinfu; Zhu, Xiulin

doi:10.1007/978-3-662-05177-1_25

Cited by 8 publications

(12 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The finding is consistent regardless of the performance evaluation metrics. The standard variant of the bag-of-words model often used in the text mining studies (Sebastiani, 2002;Yang, 1999 (Jiang et al, 2016). On the other hand, (an extension of) doc2vec…”

Section: Discussionmentioning

confidence: 99%

“…The paper represents an important contribution to the existing work on the comparative analysis of document classification performance. While the performance of different variants of the bag-of-words model is well studied (Aggarwal & Zhai, 2012;Forman, 2003;Sebastiani, 2002;Yang, 1999), the systematic comparative study of the performance of the other two document representation models is missing. Namely, the comparative studies focus on identifying the best performing classification model and/or subset of the bag-of-words features.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Influence of Feature Representation of Text on the Performance of Document Classification

2019

View full text Add to dashboard Cite

In this paper we perform a comparative analysis of three models for feature representation of text documents in the context of document classification. In particular, we consider the most often used family of models bag-of-words, recently proposed continuous space models word2vec and doc2vec, and the model based on the representation of text documents as language networks. While the bag-of-word models have been extensively used for the document classification task, the performance of the other two models for the same task have not been well understood. This is especially true for the network-based model that have been rarely considered for representation of text documents for classification. In this study, we measure the performance of the document classifiers trained using the method of random forests for features generated the three models and their variants. The results of the empirical comparison show that the commonly used bag-of-words model has performance comparable to the one obtained by the emerging continuous-space model of doc2vec. In particular, the low-dimensional variants of doc2vec generating up to 75 features are among the top-performing document representation models. The results finally point out that doc2vec shows a superior performance in the tasks of classifying large

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The Influence of Feature Representation of Text on the Performance of Document Classification

2019

View full text Add to dashboard Cite

show abstract

“…search follows from the same line of analysis used in the proof of Theorem 3. The optimization problem in (40) can be decomposed into independent subproblems (41) that can be solved in parallel: (41) where Bp t = (B k p t ) K k=1 . As we have just shown, the numerator function and the denominator function in (41) is concave and linear, respectively, so the optimization problem in (41) is a fractional programming problem and can be solved by the Dinkelbach's algorithm [33, Algorithm 5]: given λ t,τ k (λ t,0 k can be set to 0), the following optimization problem in iteration τ + 1 is solved:…”

Section: B Energy Efficiency Maximization In Massive Mimo Systemsmentioning

confidence: 99%

“…(42a) wherer k (p k ; p t ) is the numerator function in (41). The variable λ t,τ k is then updated in iteration τ + 1 as…”

Section: B Energy Efficiency Maximization In Massive Mimo Systemsmentioning

confidence: 99%

A Unified Successive Pseudoconvex Approximation Framework

Yang

Pesavento

2017

IEEE Trans. Signal Process.

169

View full text Add to dashboard Cite

Abstract-In this paper, we propose a successive pseudo-convex approximation algorithm to efficiently compute stationary points for a large class of possibly nonconvex optimization problems. The stationary points are obtained by solving a sequence of successively refined approximate problems, each of which is much easier to solve than the original problem. To achieve convergence, the approximate problem only needs to exhibit a weak form of convexity, namely, pseudo-convexity. We show that the proposed framework not only includes as special cases a number of existing methods, for example, the gradient method and the Jacobi algorithm, but also leads to new algorithms which enjoy easier implementation and faster convergence speed. We also propose a novel line search method for nondifferentiable optimization problems, which is carried out over a properly constructed differentiable function with the benefit of a simplified implementation as compared to state-of-the-art line search techniques that directly operate on the original nondifferentiable objective function. The advantages of the proposed algorithm are shown, both theoretically and numerically, by several example applications, namely, MIMO broadcast channel capacity computation, energy efficiency maximization in massive MIMO systems and LASSO in sparse signal recovery.

show abstract

“…Many studies reveal that the productivity of flares or CMEs in an AR is closely related to the complexity of the magnetic field (e.g., Zirin & Liggett 1987;Wang et al 1991;Liu et al 2016). Magnetic field evolution, such as flux emergence and cancelation (e.g., Zirin & Wang 1993;Schmieder et al 1997;Chen & Shibata 2000;Zhang et al 2001;Louis et al 2015), rapid shearing motion around the polarity inversion lines (PILs; e.g., Meunier & Kosovichev 2003;Yang et al 2004;Shimizu et al 2014), and sunspot rotation (e.g., Zhang et al 2007;Yan et al 2009Yan et al , 2015Yan et al , 2016Vemareddy et al 2016), is crucial for the occurrence of flares or CMEs. However, the role of magnetic field structure and its evolution in the production of WLFs is still poorly understood.…”

Section: Introductionmentioning

confidence: 99%

Investigation of White-light Emission in Circular-ribbon Flares

Song

2018

ApJ

View full text Add to dashboard Cite

Using observations by the Solar Dynamics Observatory from June 2010 to December 2017, we have performed the first statistical investigation of circularribbon flares (CFs) and examined the white-light emission in these CFs. We find 90 CFs occurring in 36 active regions (ARs), including 8 X-class, 34 M-class, 48 C-and B-class flares. The occurrence rate of white-light flares (WLFs) is 100% (8/8) for X-class CFs, ∼62% (21/34) for M-class CFs, and ∼8% (4/48) for Cand B-class CFs. Sometimes we observe several CFs in a single AR, and nearly all of them are WLFs. Compared to normal CFs, CFs with white-light enhancement tend to have a shorter duration, smaller size, stronger electric current and more complicated magnetic field. We find that for X-class WLFs, the whitelight enhancement is positively correlated with the flare class, implying that the white-light enhancement is largely determined by the amount of released energy. However, there is no such correlation for M-and C-class WLFs, suggesting that other factors such as the time scale, spatial scale and magnetic field complexity may play important roles in the generation of white-light emission if the released energy is not high enough.

show abstract

Y

Cited by 8 publications

References 0 publications

The Influence of Feature Representation of Text on the Performance of Document Classification

The Influence of Feature Representation of Text on the Performance of Document Classification

A Unified Successive Pseudoconvex Approximation Framework

Investigation of White-light Emission in Circular-ribbon Flares

Contact Info

Product

Resources

About