Search

The problem of probabilistic topic modeling is as follows. Given a collection of text documents, find the conditional distribution over topics for each document and the conditional distribution over words (or terms) for each topic. Log-likelihood maximization is used to solve this problem. The problem generally has an infinite set of solutions and is ill-posed according to Hadamard. In the framework of Additive Regularization of Topic Models (ARTM), a weighted sum of regularization criteria is added to the main log-likelihood criterion. The numerical method for solving this optimization problem is a kind of an iterative EM-algorithm written in a general form for an arbitrary smooth regularizer as well as for a linear combination of smooth regularizers. This paper studies the problem of convergence of the EM iterative process. Sufficient conditions are obtained for the convergence to a stationary point of the regularized log-likelihood. The constraints imposed on the regularizer are not too restrictive. We give their interpretations from the point of view of the practical implementation of the algorithm. A modification of the algorithm is proposed that improves the convergence without additional time and memory costs. Experiments on a news text collection have shown that our modification both accelerates the convergence and improves the value of the criterion to be optimized. [ABSTRACT FROM AUTHOR]

At present, in the mainstream sentiment analysis methods represented by the Support Vector Machine, the vocabulary and the latent semantic information involved in the text are not well considered, and sentiment analysis of text is dependent overly on the statistics of sentiment words. Thus, a Fisher kernel function based on Probabilistic Latent Semantic Analysis is proposed in this paper for sentiment analysis by Support Vector Machine. The Fisher kernel function based on the model is derived from the Probabilistic Latent Semantic Analysis model. By means of this method, latent semantic information involving the probability characteristics can be used as the classification characteristics, along with the improvement of the effect of classification for support vector machine, and the problem of ignoring the latent semantic characteristics in text sentiment analysis can be addressed. The results show that the effect of the method proposed in this paper, compared with the comparison method, is obviously improved. [ABSTRACT FROM AUTHOR]

Hyperspectral unmixing (HU) is an important task for remotely sensed hyperspectral (HS) data exploitation. It comprises the identification of pure spectral signatures (endmembers) and their corresponding fractional abundances in each pixel of the HS data cube. Several methods have been developed for (semi-) supervised and automatic identification of endmembers and abundances. Recently, the statistical dual-depth sparse probabilistic latent semantic analysis (DEpLSA) method has been developed to tackle the HU problem as a latent topic-based approach in which both endmembers and abundances can be simultaneously estimated according to the semantics encapsulated by the latent topic space. However, statistical models usually lead to computationally demanding algorithms and the computational time of the DEpLSA is often too high for practical use, in particular, when the dimensionality of the HS data cube is large. In order to mitigate this limitation, this article resorts to graphical processing units (GPUs) to provide a new parallel version of the DEpLSA, developed using the NVidia compute device unified architecture. Our experimental results, conducted using four well-known HS datasets and two different GPU architectures (GTX 1080 and Tesla P100), show that our parallel versions of the DEpLSA and the traditional pLSA approach can provide accurate HU results fast enough for practical use, accelerating the corresponding serial versions in at least 30x in the GTX 1080 and up to 147x in the Tesla P100 GPU, which are quite significant acceleration factors that increase with the image size, thus allowing for the possibility of the fast processing of massive HS data repositories. [ABSTRACT FROM AUTHOR]

The generative semantic nature of probabilistic topic models has recently shown encouraging results within the remote sensing image fusion field when conducting land cover categorization. However, standard topic models have not yet been adapted to the inherent complexity of remotely sensed data, which eventually may limit their resulting performance. In this scenario, this paper presents a new topic-based image fusion framework, specially designed to fuse synthetic aperture radar (SAR) and multispectral imaging (MSI) data for unsupervised land cover categorization tasks. Specifically, we initially propose a hierarchical multi-modal probabilistic latent semantic analysis (HMpLSA) model that takes advantage of two different vocabulary modalities, as well as two different levels of topics, in order to effectively uncover intersensor semantic patterns. Then, we define an SAR and MSI data fusion framework based on HMpLSA in order to perform unsupervised land cover categorization. Our experiments, conducted using three different SAR and MSI data sets, reveal that the proposed approach is able to provide competitive advantages with respect to standard clustering methods and topic models, as well as several multimodal topic model variants available in the literature. [ABSTRACT FROM AUTHOR]

At present, in the mainstream sentiment analysis methods represented by the Support Vector Machine, the vocabulary and the latent semantic information involved in the text are not well considered, and sentiment analysis of text is dependent overly on the statistics of sentiment words. Thus, a Fisher kernel function based on Probabilistic Latent Semantic Analysis is proposed in this paper for sentiment analysis by Support Vector Machine. The Fisher kernel function based on the model is derived from the Probabilistic Latent Semantic Analysis model. By means of this method, latent semantic information involving the probability characteristics can be used as the classification characteristics, along with the improvement of the effect of classification for support vector machine, and the problem of ignoring the latent semantic characteristics in text sentiment analysis can be addressed. The results show that the effect of the method proposed in this paper, compared with the comparison method, is obviously improved.

This paper presents a novel approach for spectral unmixing of remotely sensed hyperspectral data. It exploits probabilistic latent topics in order to take advantage of the semantics pervading the latent topic space when identifying spectral signatures and estimating fractional abundances from hyperspectral images. Despite the contrasted potential of topic models to uncover image semantics, they have been merely used in hyperspectral unmixing as a straightforward data decomposition process. This limits their actual capabilities to provide semantic representations of the spectral data. The proposed model, called dual-depth sparse probabilistic latent semantic analysis (DEpLSA), makes use of two different levels of topics to exploit the semantic patterns extracted from the initial spectral space in order to relieve the ill-posed nature of the unmixing problem. In other words, DEpLSA defines a first level of deep topics to capture the semantic representations of the spectra, and a second level of restricted topics to estimate endmembers and abundances over this semantic space. An experimental comparison in conducted using the two standard topic models and the seven state-of-the-art unmixing methods available in the literature. Our experiments, conducted using four different hyperspectral images, reveal that the proposed approach is able to provide competitive advantages over available unmixing approaches. [ABSTRACT FROM AUTHOR]

Probabilistic topic models have recently shown a great potential in the remote sensing image fusion field, which is particularly helpful in land-cover categorization tasks. This letter first studies the application of probabilistic latent semantic analysis (pLSA) and latent Dirichlet allocation to remote sensing synthetic aperture radar (SAR) and multispectral imaging (MSI) unsupervised land-cover categorization. Then, a novel pLSA-based image fusion approach is presented, which pursues to uncover multimodal feature patterns from SAR and MSI data in order to effectively fuse and categorize Sentinel-1 and Sentinel-2 remotely sensed data. Experiments conducted over two different data sets reveal the advantages of the proposed approach for unsupervised land-cover categorization tasks. [ABSTRACT FROM AUTHOR]

Clustering has always been one of the most challenging tasks in optical remote-sensing image (ORSI) processing, as a result of the intrinsic complexity of the distribution of the ground objects. The Gaussian mixture model (GMM), as a traditional, effective clustering method, has been widely applied. However, the traditional model does not take the spatial information into consideration. To solve the problem, a new model named the spatial Gaussian mixture model (SGMM) is proposed for ORSI clustering. The SGMM can incorporate the spatial information by generating spatial windows around pixels. An estimation algorithm based on expectation–maximization (EM) is also developed to estimate the parameters of the SGMM. The relationships between the SGMM/GMM and the SGMM/probabilistic latent semantic analysis (PLSA) are analyzed theoretically. The proposed SGMM can be considered to be an extension of the GMM and a continuous version of PLSA. In addition, two methods based on the SGMM are proposed to infer the cluster labels of the pixels. One method is based on the maximum likelihood rule, and is called SGMM-MLR, while the other method combines the SGMM and conditional random fields (CRF), and is called SGMM-CRF. The experimental results with three remote-sensing images show that the proposed clustering method based on the SGMM can improve the performance of clustering for ORSIs, compared to $k$-means, fuzzy c-means (FCM), and the GMM. It is also able to acquire a better performance than the latest cluster methods with spatial information, such as kernel weighted fuzzy local information c-means (KWFLICM), and the GMM coupled with CRF. [ABSTRACT FROM PUBLISHER]

Due to the complex arrangements of the ground objects in high spatial resolution (HSR) imagery scenes, HSR imagery scene classification is a challenging task, which is aimed at bridging the semantic gap between the low-level features and the high-level semantic concepts. A combination of multiple complementary features for HSR imagery scene classification is considered a potential way to improve the performance. However, the different types of features have different characteristics, and how to fuse the different types of features is a classic problem. In this paper, a Dirichlet-derived multiple topic model (DMTM) is proposed to fuse heterogeneous features at a topic level for HSR imagery scene classification. An efficient algorithm based on a variational expectation–maximization framework is developed to infer the DMTM and estimate the parameters of the DMTM. The proposed DMTM scene classification method is able to incorporate different types of features with different characteristics, no matter whether these features are local or global, discrete or continuous. Meanwhile, the proposed DMTM can also reduce the dimension of the features representing the HSR images. In our experiments, three types of heterogeneous features, i.e., the local spectral feature, the local structural feature, and the global textural feature, were employed. The experimental results with three different HSR imagery data sets show that the three types of features are complementary. In addition, the proposed DMTM is able to reduce the dimension of the features representing the HSR images, to fuse the different types of features efficiently, and to improve the performance of the scene classification over that of other scene classification algorithms based on spatial pyramid matching, probabilistic latent semantic analysis, and latent Dirichlet allocation. [ABSTRACT FROM PUBLISHER]

This paper presents a novel Content-Based Video Retrieval approach in order to cope with the semantic gap challenge by means of latent topics. Firstly, a supervised topic model is proposed to transform the classical retrieval approach into a class discovery problem. Subsequently, a new probabilistic ranking function is deduced from that model to tackle the semantic gap between low-level features and high-level concepts. Finally, a short-term relevance feedback scheme is defined where queries can be initialised with samples from inside or outside the database. Several retrieval simulations have been carried out using three databases and seven different ranking functions to test the performance of the presented approach. Experiments revealed that the proposed ranking function is able to provide a competitive advantage within the content-based retrieval field. [ABSTRACT FROM AUTHOR]

Scene classification has been proved to be an effective method for high spatial resolution (HSR) remote sensing image semantic interpretation. The probabilistic topic model (PTM) has been successfully applied to natural scenes by utilizing a single feature (e.g., the spectral feature); however, it is inadequate for HSR images due to the complex structure of the land-cover classes. Although several studies have investigated techniques that combine multiple features, the different features are usually quantized after simple concatenation (CAT-PTM). Unfortunately, due to the inadequate fusion capacity of \boldsymbolk-means clustering, the words of the visual dictionary obtained by CAT-PTM are highly correlated. In this paper, a semantic allocation level (SAL) multifeature fusion strategy based on PTM, namely, SAL-PTM (SAL-pLSA and SAL-LDA) for HSR imagery is proposed. In SAL-PTM: 1) the complementary spectral, texture, and scale-invariant-feature-transform features are effectively combined; 2) the three features are extracted and quantized separately by \boldsymbolk-means clustering, which can provide appropriate low-level feature descriptions for the semantic representations; and 3)the latent semantic allocations of the three features are captured separately by PTM, which follows the core idea of PTM-based scene classification. The probabilistic latent semantic analysis (pLSA) and latent Dirichlet allocation (LDA) models were compared to test the effect of different PTMs for HSR imagery. A U.S. Geological Survey data set and the UC Merced data set were utilized to evaluate SAL-PTM in comparison with the conventional methods. The experimental results confirmed that SAL-PTM is superior to the single-feature methods and CAT-PTM in the scene classification of HSR imagery. [ABSTRACT FROM PUBLISHER]

The interest in image annotation and recommendation has been increased due to the ever rising amount of data uploaded to the web. Despite the many efforts undertaken so far, accuracy or efficiency still remain open problems. Here, a complete image annotation and tourism recommender system is proposed. It is based on the probabilistic latent semantic analysis (PLSA) and hypergraph ranking, exploiting the visual attributes of the images and the semantic information found in image tags and geo-tags. In particular, semantic image annotation resorts to the PLSA, exploiting the textual information in image tags. It is further complemented by visual annotation based on visual image content classification. Tourist destinations, strongly related to a query image, are recommended using hypergraph ranking enhanced by enforcing group sparsity constraints. Experiments were conducted on a large image dataset of Greek sites collected from Flickr. The experimental results demonstrate the merits of the proposed model. Semantic image annotation by means of the PLSA has achieved an average precision of 92% at 10% recall. The accuracy of content-based image classification is 82, 6%. An average precision of 92% is measured at 1% recall for tourism recommendation. [ABSTRACT FROM AUTHOR]

Recent research trends in Content-based Video Retrieval have shown topic models as an effective tool to deal with the semantic gap challenge. In this scenario, this paper has a dual target: (1) it is aimed at studying how the use of different topic models (pLSA, LDA and FSTM) affects video retrieval performance; (2) a novel incremental topic model (IpLSA) is presented in order to cope with incremental scenarios in an effective and efficient way. A comprehensive comparison among these four topic models using two different retrieval systems and two reference benchmarking video databases is provided. Experiments revealed that pLSA is the best model in sparse conditions, LDA tend to outperform the rest of the models in a dense space and IpLSA is able to work properly in both cases. [ABSTRACT FROM AUTHOR]

Image annotation is a process of finding appropriate semantic labels for images in order to obtain a more convenient way for indexing and searching images on theWeb. This article proposes a novel method for image annotation based on combining feature-word distributions, which map from visual space to word space, and word-topic distributions, which form a structure to capture label relationships for annotation. We refer to this type of model as Feature-Word-Topic models. The introduction of topics allows us to efficiently take word associations, such as {ocean, fish, coral} or {desert, sand, cactus}, into account for image annotation. Unlike previous topic-based methods, we do not consider topics as joint distributions of words and visual features, but as distributions of words only. Feature-word distributions are utilized to define weights in computation of topic distributions for annotation. By doing so, topic models in text mining can be applied directly in our method. Our Feature-word-topic model, which exploits Gaussian Mixtures for feature-word distributions, and probabilistic Latent Semantic Analysis (pLSA) for word-topic distributions, shows that our method is able to obtain promising results in image annotation and retrieval. [ABSTRACT FROM AUTHOR]

Abstract: Focusing on the problem of localized content-based image retrieval, based on probabilistic latent semantic analysis (PLSA) and transductive support vector machine (TSVM), a novel semi-supervised multi-instance learning (SSMIL) algorithm is proposed, where a bag corresponds to an image and an instance corresponds to the low-level visual features of a segmented region. In order to convert an MIL problem into a standard supervised learning problem, first, all the instances in training bags be clustered by -Means method, and regards each cluster center as “visual-word” to build a visual vocabulary. Second, according to the distance between “visual-word” and instance, a fuzzy membership function is defined to establish a fuzzy term-document matrix, then use PLSA method to obtain bag’s (image’s) latent topic models, which can convert every bag to a single sample. Finally, in order to use the unlabeled images to improve retrieval accuracy, using semi-supervised TSVM to train classifiers. Experimental results on the COREL data sets show that the proposed method, named PLSA-SSMIL, is robust, and its performance is superior to other key existing MIL algorithms. [Copyright &y& Elsevier]

Segmentation of cells/nuclei is a challenging problem in 2-D histological and cytological images. Although a large number of algorithms have been proposed, newer efforts continue to be devoted to investigate robust models that could have high level of adaptability with regard to considerable amount of image variability. In this paper, we propose a multiclassification conditional random fields (CRFs) model using a combination of low-level cues (bottom–up) and high-level contextual information (top–down) for separating nuclei from the background. In our approach, the contextual information is extracted by an unsupervised topic discovery process, which efficiently helps to suppress segmentation errors caused by intensity inhomogeneity and variable chromatin texture. In addition, we propose a multilayer CRF, an extension of the traditional single-layer CRF, to handle high-dimensional dataset obtained through spectral microscopy, which provides combined benefits of spectroscopy and imaging microscopy, resulting in the ability to acquire spectral images of microscopic specimen. The approach is evaluated with color images, as well as spectral images. The overall accuracy of the proposed segmentation algorithm reaches 95% when applying multilayer CRF model to the spectral microscopy dataset. Experiments also show that our method outperforms seeded watershed, a widely used algorithm for cell segmentation. [ABSTRACT FROM PUBLISHER]

Given its importance, the problem of selecting the right site for a service entity has attracted great attention in the literature. However, due to its complexity, the quantification of the interrelationships between the service site and its nearby business types is still a challenging task. To this end, in this paper, we propose a novel joint learning scheme for service site selection. This scheme employs both the Probabilistic Latent Semantic Analysis (PLSA) on the Geographical Information System (GIS) data and the partitional clustering on the service performance data. A case study for bank branch selection is provided to demonstrate the usefulness of our method. Finally, based on the joint learning scheme, we present a conceptual framework for the complete procedure of service site selection with a particular emphasis on the GIS enabled network analysis. [ABSTRACT FROM AUTHOR]