Previous studies on Japanese fricatives and affricates revealed that [s] and [ts] are classified by variables in the time domain (Yamakawa et al., 2012), whereas [ts] and [ʨ] as well as [s] and [ʨ] are classified by variables in the spectral domain (Yamakawa & Amano, 2011). To gain an integrated perspective on these findings, this study examined whether [s] [ts], and [ʨ] can be classified by a ...

This study assessed 96 law school students’ preferences for online, hybrid, or traditional learning environments, and their reasons for these preferences, learning strategies, and motivational orientations. A discriminant analysis revealed that non-traditional learning environment familiarity, self-efficacy, and employment status were the strongest predictors of preferences for non-traditional ...

Figure 3. Position of each case on linear discriminant 1 (LD1) versus linear discriminant 2 (LD2). Spectral moments analysis has been shown to be effective in deriving acoustic features for classifying voiceless stop release bursts [1], and is an analysis method that has commonly been cited in the clinical phonetics literature dealing with children’s disordered speech. In this study, we compare...

The problem of total participation in asynchronous meetings convened with distributed group support systems has been noted frequently but investigated rarely. This paper describes a portion of a large field study using the distributed group support system WebWide Participation in which explanations for meeting involvement (and noninvolvement) were explored. In particular, three WebWide meetings...

In this paper, a novel nonlinear discriminant analysis is proposed. Experimental results show that the new method provides state of the art performance when combined with LSVM in terms of training time and accuracy.

在近十年來所發展出的自動語音辨識(automatic speech recognition, ASR)技術中,仍 有許多研究者嘗試僅藉由前端處理來產生具有鑑別性的語音特徵,而獨立於後端模型訓 練與分類器特性。本論文即在此思維下提出嶄新的鑑別式特徵轉換方法,稱為普遍化相 似度比率鑑別分析(generalized likelihood ratio discriminant analysis, GLRDA),其旨在利 用相似度比率檢驗(likelihood ratio test)的概念尋求一個維度較低的特徵空間。在此子空 間中,我們不僅考慮了全體資料的異方差性(heteroscedasticity),即所有類別之共變異矩 陣可被彈性地視為相異,並且在分類上,因著我們也將類別間最混淆之情況(由虛無假 設(null hypothesis)所描述)的發生率降至最低,而達到有助於分類正確率提升的效...

The usefulness of dynamic formant properties for speaker discrimination was demonstrated on English, mostly exploiting long vowels or diphthongs, characterized both by actual measures along the formant tracks and coefficients from polynomial regression on these tracks. This study applies this paradigm to Czech, using less tightly controlled (more forensically realistic) material and taking into...

We provide a general theoretical framework to derive Bernstein-von Mises theorems for matrix functionals. The conditions on functionals and priors are explicit and easy to check. Results are obtained for various functionals including entries of covariance matrix, entries of precision matrix, quadratic forms, log-determinant, eigenvalues in the Bayesian Gaussian covariance/precision matrix estim...

To begin, we first use B to project all the training data into a low-dimensional discriminant space. Let Σw and μk denote the within-class pooled covariance matrix and the mean of class k respectively for all the projected data. When given a new observation x, we assume that it is draw from a normal distribution with mean given by some μk and covariance given by Σ B w . We then assign x ? to th...

Using in situ hyperspectral measurements collected in the Sierra Nevada Mountains in California, we discriminate six species of conifer trees using a recent, nonparametric statistics technique known as penalized discriminant analysis (PDA). A classification accuracy of 76% is obtained. Our emphasis is on providing an intuitive, geometric description of PDA that makes the advantages of penalizat...