Huge amounts of dimensionality reduction methods [12] are proposed in the past decades. Wang et al. [13] propose capped p -norm LDA method to find optimal subspace. Shen et al. [14] propose a generalized least-squares approach regularized with graph embedding to keep local structure and ...
Empty Cell [105] PCA + LDA CUFSF CITE 500:694 99% Empty Cell [30] NN, Chi-square CUFS Gabor Shape 306:300 99% Empty Cell [101] Weighted Chi-square CUFS EUCLBP 78:233 94% Empty Cell [29] NN, Chi-square CUFS HAOG 306:300 100% Empty Cell [30] NN, Chi-square CUFSF Gabor Sha...
Benouareth A (2021) An efficient face recognition approach combining likelihood-based sufficient dimension reduction and LDA. Multmed Tool Appl 80:1457–1486 Article Google Scholar Qaraei M, Abbaasi S, Shirazi KG (2021) Randomized non-linear PCA networks. Inf Sci 545:241–253 Article MathSci...
58) What is the difference between LDA and PCA for dimensionality reduction? Both LDA and PCA are linear transformation techniques: LDA is a supervised whereas PCA is unsupervised – PCA ignores class labels. We can picture PCA as a technique that finds the directions of maximal variance. In ...
This alternative scheme was put forward by authors of the LDAK9 model and have been examined carefully in their subsequent works. In simulations, we observed that indeed, HEELS can be biased under mis-specified models, and that the amount of bias varies with the strength of MAF/LD dependency...
The result demonstrates that dimensionality reduction using the topic model is beneficial for image classification. In this paper, we follow the second extension and construct Local-Class–Shared–Topic LDA (lcst–LDA). In our model, the random variables and the model parameters are represented by ...
PCA, LDA, and their variants [5, 6] are not able to reveal the underlying non-linear [3, 4] structure of the face data. Recently, many manifold learning-based algorithms with locality preserving abilities have been presented. Among them, isometric feature mapping (ISOMAP) [7], locally ...
The second step reduces the histograms dimension using the whitened linear discriminate analysis (WLDA), which gives scale-invariant DMS-BSIF features. Finally, the K-NN classifier was used to identify the input ear images. It is practically hard to produce all specific characteristics using an ...
FDA-L1 [29] provides a robust alternative algorithm to the traditional LDA algorithm that is not limited by SSS problems. On the basis of the PCA-L1 algorithm, Li et al. proposed 2DPCA [30] algorithm based on the L1 norm. In reference [31], the 2DLPP-L1 and 2DDLPP-L1 are ...
A similar idea to deal with text image information was applied by Yuan et al. in a very recent work [17]. They propose a new framework to address the diversity-induced image retrieval problem that applies latent Dirichlet allocation (LDA) to the initial retrieval results to discover some ...