In this paper, we use non-negative matrix factorization (NMF) to analyze the data from stock market. By using the multiplicative update rules algorithm, we decompose the data matrix V of the daily closing prices of th...
详细信息
ISBN:
(纸本)9781424441334
In this paper, we use non-negative matrix factorization (NMF) to analyze the data from stock market. By using the multiplicative update rules algorithm, we decompose the data matrix V of the daily closing prices of the 40 stocks, of which the Shenzhen component index is made up, into two matrices W and H, in which the columns of W correlate to the underlying trends. In addition, the Euclidean distance between the 40 stocks and the underlying forces is constructed. By means of the Kmeans routine in MATLAB, the 40 stocks are classified into different clusters with the center of the underlying forces, which can be finished automatically by MATLAB. Finally, the properties of these clusters are studied.
The computer vision problem of face classification under several ambient and unfavorable conditions is considered in this study. Changes in expression, different lighting conditions and occlusions are the relevant fac...
详细信息
ISBN:
(纸本)3540000119
The computer vision problem of face classification under several ambient and unfavorable conditions is considered in this study. Changes in expression, different lighting conditions and occlusions are the relevant factors that are studied, in this present contribution. non-negative matrix factorization (NMF) technique is introduced in the context of face classification and a direct comparison with Principal Component Analysis (PCA) is also analyzed. Two leading techniques in face recognition are also considered in this study noticing that NMF is able to improve these. techniques. when a high dimensional feature space is used. Finally, different distance metrics (L1, L2 and correlation) are evaluated in the feature space defined by NMF in order to determine the best one for this specific problem. Experiments demonstrate that the correlation is the most suitable metric for this problem.
In many event detection applications, training data may contain tags with multiple, simultaneous events. This is particularly likely when the definition of "event" is broad and includes events that can persi...
详细信息
ISBN:
(纸本)9781509041176
In many event detection applications, training data may contain tags with multiple, simultaneous events. This is particularly likely when the definition of "event" is broad and includes events that can persist for an extended period of time. Decomposing a mixed signal into signals corresponding to individual events is non-trivial. In this paper, we propose a non-negative matrix factorization (NMF) method that generates independent dictionaries for different events from training data with overlapping events. The proposed method adds a mask matrix into the regularization term in conventional NMF approaches. This mask matrix captures known event labels in the training data, so that only related dictionary terms are updated during iteration. The effectiveness of the proposed approach is evaluated using both synthetic and real data.
We present a probabilistic framework for efficient non-negative matrix factorization of discrete (count/binary) data with side-information. The side-information is given as a multi-level structure, taxonomy, or ontolo...
详细信息
We present a probabilistic framework for efficient non-negative matrix factorization of discrete (count/binary) data with side-information. The side-information is given as a multi-level structure, taxonomy, or ontology, with nodes at each level being categorical-valued observations. For example, when modeling documents with a two-level side-information (documents being at level-zero), level-one may represent (one or more) authors associated with each document and level-two may represent affiliations of each author. The model easily generalizes to more than two levels (or taxonomy/ontology of arbitrary depth). Our model can learn embeddings of entities present at each level in the data/side-information hierarchy (e.g., documents, authors, affiliations, in the previous example), with appropriate sharing of information across levels. The model also enjoys full local conjugacy, facilitating efficient Gibbs sampling for model inference. Inference cost scales in the number of non-zero entries in the data matrix, which is especially appealing for real-world massive but sparse matrices. We demonstrate the effectiveness of the model on several real-world data sets.
As the numbers of document in Chinese and oriental languages increased in recent years, it becomes increasingly important to develop oriental-language-document filtering systems. In oriental language documents, the cl...
详细信息
ISBN:
(纸本)1932415270
As the numbers of document in Chinese and oriental languages increased in recent years, it becomes increasingly important to develop oriental-language-document filtering systems. In oriental language documents, the classical problems of synonymy and polysemy still exists, so the filtering method based on the latent semantic indexing (LSI), which represent documents by semantic relations between words, perform better than other methods which represent documents just by words. non-negative matrix factorization (NMF), another method for dimensionality reduction and distinguished from LSI by its non-negativity constraints, has supervised LSI in many other fields, such as English-document clustering and classifying etc. In this paper, we propose a new method based on NMF to obtain topic profiles from the set of sample documents, and use them for document filtering. The experimental results show that the new method is better than a highly effective method based on LSI in filtering the oriental language documents.
We introduce a novel approach for noise-robust feature extraction in speech recognition, based on non-negative matrix factorization (NMF). While NMF has previously been used for speech denoising and speaker separation...
详细信息
ISBN:
(纸本)9781424442966
We introduce a novel approach for noise-robust feature extraction in speech recognition, based on non-negative matrix factorization (NMF). While NMF has previously been used for speech denoising and speaker separation, we directly extract time-varying features from the NMF output. To this end we extend basic unsupervised NMF to a hybrid supervised/unsupervised algorithm. We present a Dynamic Bayesian Network (DBN) architecture that can exploit these features in a Tandem manner together with the maximum likelihood phoneme estimate of a bidirectional long short-term memory (BLSTM) recurrent neural network. We show that addition of NMF features to spelling recognition systems can increase word accuracy by up to 7% absolute in a noisy car environment.
nonnegativematrixfactorization (NMF) is a popular method for source separation. In this paper, an alternating direction method of multipliers (ADMM) for NMF is studied, which deals with the NMF problem using the cos...
详细信息
ISBN:
(纸本)9789881476807
nonnegativematrixfactorization (NMF) is a popular method for source separation. In this paper, an alternating direction method of multipliers (ADMM) for NMF is studied, which deals with the NMF problem using the cost function of beta-divergence. Our study shows that this algorithm outperforms state-of-the-art algorithms on synthetic data sets, but it presents unstable behavior and low accuracy on real data sets. Therefore, we propose two different stable ADMM algorithms for NMF to solve this problem. They differ slightly in the multiplicative factor utilized in the update rules. One algorithm is to adapt the step size to guarantee the convergence while the other minimizes the beta-divergence with a pivot element weighting iterative method (PEWI). Experimental results demonstrate that the proposed algorithms are more stable and accurate. Particularly, PEWI based ADMM shows superior performance in the source separation task.
In traditional supervised learning, the model is trained through a large number of known labels. In reality, the amount of data increases rapidly, and the cost of manual annotation is high. Besides, under the restrict...
详细信息
In traditional supervised learning, the model is trained through a large number of known labels. In reality, the amount of data increases rapidly, and the cost of manual annotation is high. Besides, under the restriction of privacy protection, it is infeasible to obtain labels of a large number of samples. However, the proportion of samples in a certain class can be easily obtained Thanks to label proportional learning, we can obtain the instance-level classifier merely based on the proportional information. In addition to this, label proportional learning with minor instance-level labeling can potentially improve the accuracy of the model. Hence, in this paper, we propose an improved model based on non-negative matrix factorization to solve semi-supervised label proportional learning, called semi-supervised proportion matrixfactorization (SPMF). In order to effectively predict car default customers, our method leverage a small number of known labels and the remaining sample label proportional information to construct a classification model. In our experiment, customers defaults can be predicted and high prediction accuracy can be guaranteed. This study is potentially useful for Internet auto financial institutions to properly avoid default risks and establish a more reliable credit rating mechanism. Particularly, our approach provides an alternative way to effectively solve default customer identification under the privacy protection constraint. (C) 2020 The Authors. Published by Elsevier B.V.
In this paper we study the effect of regularization on clustering results provided by non-negative matrix factorization (NMF). Different kinds of regularization terms were previously added to the NMF objective functio...
详细信息
ISBN:
(纸本)9783319126401
In this paper we study the effect of regularization on clustering results provided by non-negative matrix factorization (NMF). Different kinds of regularization terms were previously added to the NMF objective function in order to produce sparser results and thus to obtain a more qualitative partition of data. We would like to propose the general framework for regularized NMF based on Schatten p-norms. Experimental results show the effectiveness of our approach on different data sets.
The constrained low-rank and sparse matrix decomposition (CLSMD) method ignores the temporal continuity between adjacent speech frames in the process of speech enhancement, resulting in a sparse matrix generated by de...
详细信息
ISBN:
(纸本)9781538662434
The constrained low-rank and sparse matrix decomposition (CLSMD) method ignores the temporal continuity between adjacent speech frames in the process of speech enhancement, resulting in a sparse matrix generated by decomposition with isolated discrete points. Therefore, in order to improve the noise suppression ability of the speech system and improve the enhanced speech quality and intelligibility, this paper proposes a speech enhancement method based on Temporal continuity Constraint for non-negative Low-rank and Sparse matrix Decomposition (TCNLSMD). In this method, in addition to adding low -rank and sparse constraints, temporal continuity constraints are added. The proposed method based on the sparse matrix obtained by eigenvalue decomposition of non-negative matrices and hard-threshold function estimation, the discrete sparse matrix is reduced by adding temporal continuity constraints to reduce discrete isolated points, retaining more speech information and reducing the enhanced speech distortion. The experimental results show that under various types of noise test conditions, compared with the current mainstream speech enhancement methods, especially with NLSMD, the proposed method improve the noise suppression capability, make the residual noise less, and improve the quality of the enhanced speech.
暂无评论