Showing papers on "Probabilistic latent semantic analysis published in 2014"

A Latent Semantic Model with Convolutional-Pooling Structure for Information Retrieval

Abstract: In this paper, we propose a new latent semantic model that incorporates a convolutional-pooling structure over word sequences to learn low-dimensional, semantic vector representations for search queries and Web documents. In order to capture the rich contextual structures in a query or a document, we start with each word within a temporal context window in a word sequence to directly capture contextual features at the word n-gram level. Next, the salient word n-gram features in the word sequence are discovered by the model and are then aggregated to form a sentence-level feature vector. Finally, a non-linear transformation is applied to extract high-level semantic information to generate a continuous vector representation for the full text string. The proposed convolutional latent semantic model (CLSM) is trained on clickthrough data and is evaluated on a Web document ranking task using a large-scale, real-world data set. Results show that the proposed model effectively captures salient semantic information in queries and documents for the task while significantly outperforming previous state-of-the-art semantic models.

BTM: Topic Modeling over Short Texts

Abstract: Short texts are popular on today’s web, especially with the emergence of social media. Inferring topics from large scale short texts becomes a critical but challenging task for many content analysis tasks. Conventional topic models such as latent Dirichlet allocation (LDA) and probabilistic latent semantic analysis (PLSA) learn topics from document-level word co-occurrences by modeling each document as a mixture of topics, whose inference suffers from the sparsity of word co-occurrence patterns in short texts. In this paper, we propose a novel way for short text topic modeling, referred as biterm topic model (BTM) . BTM learns topics by directly modeling the generation of word co-occurrence patterns (i.e., biterms) in the corpus, making the inference effective with the rich corpus-level information. To cope with large scale short text data, we further introduce two online algorithms for BTM for efficient topic learning. Experiments on real-word short text collections show that BTM can discover more prominent and coherent topics, and significantly outperform the state-of-the-art baselines. We also demonstrate the appealing performance of the two online BTM algorithms on both time efficiency and topic learning.

Latent Regression Forest: Structured Estimation of 3D Articulated Hand Posture

Abstract: In this paper we present the Latent Regression Forest (LRF), a novel framework for real-time, 3D hand pose estimation from a single depth image. In contrast to prior forest-based methods, which take dense pixels as input, classify them independently and then estimate joint positions afterwards, our method can be considered as a structured coarse-to-fine search, starting from the centre of mass of a point cloud until locating all the skeletal joints. The searching process is guided by a learnt Latent Tree Model which reflects the hierarchical topology of the hand. Our main contributions can be summarised as follows: (i) Learning the topology of the hand in an unsupervised, data-driven manner. (ii) A new forest-based, discriminative framework for structured search in images, as well as an error regression step to avoid error accumulation. (iii) A new multi-view hand pose dataset containing 180K annotated images from 10 different subjects. Our experiments show that the LRF out-performs state-of-the-art methods in both accuracy and efficiency.

Latent semantic sparse hashing for cross-modal similarity search

Abstract: Similarity search methods based on hashing for effective and efficient cross-modal retrieval on large-scale multimedia databases with massive text and images have attracted considerable attention. The core problem of cross-modal hashing is how to effectively construct correlation between multi-modal representations which are heterogeneous intrinsically in the process of hash function learning. Analogous to Canonical Correlation Analysis (CCA), most existing cross-modal hash methods embed the heterogeneous data into a joint abstraction space by linear projections. However, these methods fail to bridge the semantic gap more effectively, and capture high-level latent semantic information which has been proved that it can lead to better performance for image retrieval. To address these challenges, in this paper, we propose a novel Latent Semantic Sparse Hashing (LSSH) to perform cross-modal similarity search by employing Sparse Coding and Matrix Factorization. In particular, LSSH uses Sparse Coding to capture the salient structures of images, and Matrix Factorization to learn the latent concepts from text. Then the learned latent semantic features are mapped to a joint abstraction space. Moreover, an iterative strategy is applied to derive optimal solutions efficiently, and it helps LSSH to explore the correlation between multi-modal representations efficiently and automatically. Finally, the unified hashcodes are generated through the high level abstraction space by quantization. Extensive experiments on three different datasets highlight the advantage of our method under cross-modal scenarios and show that LSSH significantly outperforms several state-of-the-art methods.

Build, Compute, Critique, Repeat: Data Analysis with Latent Variable Models

Abstract: We survey latent variable models for solving data-analysis problems. A latent variable model is a probabilistic model that encodes hidden patterns in the data. We uncover these patterns from their conditional distribution and use them to summarize data and form predictions. Latent variable models are important in many fields, including computational biology, natural language processing, and social network analysis. Our perspective is that models are developed iteratively: We build a model, use it to analyze data, assess how it succeeds and fails, revise it, and repeat. We describe how new research has transformed these essential activities. First, we describe probabilistic graphical models, a language for formulating latent variable models. Second, we describe mean field variational inference, a generic algorithm for approximating conditional distributions. Third, we describe how to use our analyses to solve problems: exploring the data, forming predictions, and pointing us in the direction of improved mo...

Weakly Supervised Object Localization with Latent Category Learning

Abstract: Localizing objects in cluttered backgrounds is a challenging task in weakly supervised localization. Due to large object variations in cluttered images, objects have large ambiguity with backgrounds. However, backgrounds contain useful latent information, e.g., the sky for aeroplanes. If we can learn this latent information, object-background ambiguity can be reduced to suppress the background. In this paper, we propose the latent category learning (LCL), which is an unsupervised learning problem given only image-level class labels. Firstly, inspired by the latent semantic discovery, we use the typical probabilistic Latent Semantic Analysis (pLSA) to learn the latent categories, which can represent objects, object parts or backgrounds. Secondly, to determine which category contains the target object, we propose a category selection method evaluating each category’s discrimination. We evaluate the method on the PASCAL VOC 2007 database and ILSVRC 2013 detection challenge. On VOC 2007, the proposed method yields the annotation accuracy of 48%, which outperforms previous results by 10%. More importantly, we achieve the detection average precision of 30.9%, which improves previous results by 8% and can be competitive with the supervised deformable part model (DPM) 5.0 baseline 33.7%. On ILSVRC 2013 detection, the method yields the precision of 6.0%, which is also competitive with the DPM 5.0.

Nonparametric Bayes dynamic modelling of relational data

Abstract: Symmetric binary matrices representing relations are collected in many areas. Our focus is on dynamically evolving binary relational matrices, with interest being on inference on the relationship structure and prediction. We propose a nonparametric Bayesian dynamic model, which reduces dimensionality in characterizing the binary matrix through a lower-dimensional latent space representation, with the latent coordinates evolving in continuous time via Gaussian processes. By using a logistic mapping function from the link probability matrix space to the latent relational space, we obtain a flexible and computationally tractable formulation. Employing Polya-gamma data augmentation, an efficient Gibbs sampler is developed for posterior computation, with the dimension of the latent space automatically inferred. We provide theoretical results on flexibility of the model, and illustrate its performance via simulation experiments. We also consider an application to co-movements in world financial markets.

Scalable Link Prediction in Dynamic Networks via Non-Negative Matrix Factorization.

Abstract: We propose a scalable temporal latent space model for link prediction in dynamic social networks, where the goal is to predict links over time based on a sequence of previous graph snapshots. The model assumes that each user lies in an unobserved latent space and interactions are more likely to form between similar users in the latent space representation. In addition, the model allows each user to gradually move its position in the latent space as the network structure evolves over time. We present a global optimization algorithm to effectively infer the temporal latent space, with a quadratic convergence rate. Two alternative optimization algorithms with local and incremental updates are also proposed, allowing the model to scale to larger networks without compromising prediction accuracy. Empirically, we demonstrate that our model, when evaluated on a number of real-world dynamic networks, significantly outperforms existing approaches for temporal link prediction in terms of both scalability and predictive power.

Text classification using genetic algorithm oriented latent semantic features

Abstract: In this paper, genetic algorithm oriented latent semantic features (GALSF) are proposed to obtain better representation of documents in text classification. The proposed approach consists of feature selection and feature transformation stages. The first stage is carried out using the state-of-the-art filter-based methods. The second stage employs latent semantic indexing (LSI) empowered by genetic algorithm such that a better projection is attained using appropriate singular vectors, which are not limited to the ones corresponding to the largest singular values, unlike standard LSI approach. In this way, the singular vectors with small singular values may also be used for projection whereas the vectors with large singular values may be eliminated as well to obtain better discrimination. Experimental results demonstrate that GALSF outperforms both LSI and filter-based feature selection methods on benchmark datasets for various feature dimensions.

Biomedical text mining: State-of-the-art, open problems and future challenges

Abstract: Text is a very important type of data within the biomedical domain. For example, patient records contain large amounts of text which has been entered in a non-standardized format, consequently posing a lot of challenges to processing of such data. For the clinical doctor the written text in the medical findings is still the basis for decision making – neither images nor multimedia data. However, the steadily increasing volumes of unstructured information need machine learning approaches for data mining, i.e. text mining. This paper provides a short, concise overview of some selected text mining methods, focusing on statistical methods, i.e. Latent Semantic Analysis, Probabilistic Latent Semantic Analysis, Latent Dirichlet Allocation, Hierarchical Latent Dirichlet Allocation, Principal Component Analysis, and Support Vector Machines, along with some examples from the biomedical domain. Finally, we provide some open problems and future challenges, particularly from the clinical domain, that we expect to stimulate future research.

Bayesian Estimation of Discrete Multivariate Latent Structure Models With Structural Zeros

Abstract: In multivariate categorical data, models based on conditional independence assumptions, such as latent class models, offer efficient estimation of complex dependencies. However, Bayesian versions of latent structure models for categorical data typically do not appropriately handle impossible combinations of variables, also known as structural zeros. Allowing nonzero probability for impossible combinations results in inaccurate estimates of joint and conditional probabilities, even for feasible combinations. We present an approach for estimating posterior distributions in Bayesian latent structure models with potentially many structural zeros. The basic idea is to treat the observed data as a truncated sample from an augmented dataset, thereby allowing us to exploit the conditional independence assumptions for computational expediency. As part of the approach, we develop an algorithm for collapsing a large set of structural zero combinations into a much smaller set of disjoint marginal conditions, which sp...

Towards geospatial semantic search: exploiting latent semantic relations in geospatial data

Abstract: This paper reports our efforts to address the grand challenge of the Digital Earth vision in terms of intelligent data discovery from vast quantities of geo-referenced data. We propose an algorithm combining LSA and a Two-Tier Ranking (LSATTR) algorithm based on revised cosine similarity to build a more efficient search engine – Semantic Indexing and Ranking (SIR) – for a semantic-enabled, more effective data discovery. In addition to its ability to handle subject-based search, we propose a mechanism to combine geospatial taxonomy and Yahoo! GeoPlanet for automatic identification of location information from a spatial query and automatic filtering of datasets that are not spatially related. The metadata set, in the format of ISO19115, from NASA's SEDAC (Socio-Economic Data Application Center) is used as the corpus of SIR. Results show that our semantic search engine SIR built on LSATTR methods outperforms existing keyword-matching techniques, such as Lucene, in terms of both recall and precision. Moreover...

Latent Markov models: a review of a general framework for the analysis of longitudinal data with covariates

Abstract: We provide a comprehensive overview of latent Markov (LM) models for the analysis of longitudinal categorical data. We illustrate the general version of the LM model which includes individual covariates, and several constrained versions. Constraints make the model more parsimonious and allow us to consider and test hypotheses of interest. These constraints may be put on the conditional distribution of the response variables given the latent process (measurement model) or on the distribution of the latent process (latent model). We also illustrate in detail maximum likelihood estimation through the Expectation–Maximization algorithm, which may be efficiently implemented by recursions taken from the hidden Markov literature. We outline methods for obtaining standard errors for the parameter estimates. We also illustrate methods for selecting the number of states and for path prediction. Finally, we mention issues related to Bayesian inference of LM models. Possibilities for further developments are given among the concluding remarks.

Mixture of latent trait analyzers for model-based clustering of categorical data

Abstract: Model-based clustering methods for continuous data are well established and commonly used in a wide range of applications. However, model-based clustering methods for categorical data are less standard. Latent class analysis is a commonly used method for model-based clustering of binary data and/or categorical data, but due to an assumed local independence structure there may not be a correspondence between the estimated latent classes and groups in the population of interest. The mixture of latent trait analyzers model extends latent class analysis by assuming a model for the categorical response variables that depends on both a categorical latent class and a continuous latent trait variable; the discrete latent class accommodates group structure and the continuous latent trait accommodates dependence within these groups. Fitting the mixture of latent trait analyzers model is potentially difficult because the likelihood function involves an integral that cannot be evaluated analytically. We develop a variational approach for fitting the mixture of latent trait models and this provides an efficient model fitting strategy. The mixture of latent trait analyzers model is demonstrated on the analysis of data from the National Long Term Care Survey (NLTCS) and voting in the U.S. Congress. The model is shown to yield intuitive clustering results and it gives a much better fit than either latent class analysis or latent trait analysis alone.

Medical Image Retrieval: A Multimodal Approach.

Abstract: Medical imaging is becoming a vital component of war on cancer. Tremendous amounts of medical image data are captured and recorded in a digital format during cancer care and cancer research. Facing such an unprecedented volume of image data with heterogeneous image modalities, it is necessary to develop effective and efficient content-based medical image retrieval systems for cancer clinical practice and research. While substantial progress has been made in different areas of content-based image retrieval (CBIR) research, direct applications of existing CBIR techniques to the medical images produced unsatisfactory results, because of the unique characteristics of medical images. In this paper, we develop a new multimodal medical image retrieval approach based on the recent advances in the statistical graphic model and deep learning. Specifically, we first investigate a new extended probabilistic Latent Semantic Analysis model to integrate the visual and textual information from medical images to bridge the semantic gap. We then develop a new deep Boltzmann machine-based multimodal learning model to learn the joint density model from multimodal information in order to derive the missing modality. Experimental results with large volume of real-world medical images have shown that our new approach is a promising solution for the next-generation medical imaging indexing and retrieval system.

BayesLCA: An R Package for Bayesian Latent Class Analysis

Abstract: The BayesLCA package for R provides tools for performing latent class analysis within a Bayesian setting. Three methods for fitting the model are provided, incorporating an expectation-maximization algorithm, Gibbs sampling and a variational Bayes approximation. The article briefly outlines the methodology behind each of these techniques and discusses some of the technical difficulties associated with them. Methods to remedy these problems are also described. Visualization methods for each of these techniques are included, as well as criteria to aid model selection.

Learning and Recognition of On-Premise Signs From Weakly Labeled Street View Images

Abstract: Camera-enabled mobile devices are commonly used as interaction platforms for linking the user's virtual and physical worlds in numerous research and commercial applications, such as serving an augmented reality interface for mobile information retrieval. The various application scenarios give rise to a key technique of daily life visual object recognition. On-premise signs (OPSs), a popular form of commercial advertising, are widely used in our living life. The OPSs often exhibit great visual diversity (e.g., appearing in arbitrary size), accompanied with complex environmental conditions (e.g., foreground and background clutter). Observing that such real-world characteristics are lacking in most of the existing image data sets, in this paper, we first proposed an OPS data set, namely OPS-62, in which totally 4649 OPS images of 62 different businesses are collected from Google's Street View. Further, for addressing the problem of real-world OPS learning and recognition, we developed a probabilistic framework based on the distributional clustering, in which we proposed to exploit the distributional information of each visual feature (the distribution of its associated OPS labels) as a reliable selection criterion for building discriminative OPS models. Experiments on the OPS-62 data set demonstrated the outperformance of our approach over the state-of-the-art probabilistic latent semantic analysis models for more accurate recognitions and less false alarms, with a significant 151.28% relative improvement in the average recognition rate. Meanwhile, our approach is simple, linear, and can be executed in a parallel fashion, making it practical and scalable for large-scale multimedia applications.

Multi-modal Mutual Topic Reinforce Modeling for Cross-media Retrieval

Abstract: As an important and challenging problem in the multimedia area, multi-modal data understanding aims to explore the intrinsic semantic information across different modalities in a collaborative manner. To address this problem, a possible solution is to effectively and adaptively capture the common cross-modal semantic information by modeling the inherent correlations between the latent topics from different modalities. Motivated by this task, we propose a supervised multi-modal mutual topic reinforce modeling (M$^3$R) approach, which seeks to build a joint cross-modal probabilistic graphical model for discovering the mutually consistent semantic topics via appropriate interactions between model factors (e.g., categories, latent topics and observed multi-modal data). In principle, M$^3$R is capable of simultaneously accomplishing the following two learning tasks: 1) modality-specific (e.g., image-specific or text-specific ) latent topic learning; and 2) cross-modal mutual topic consistency learning. By investigating the cross-modal topic-related distribution information, M$^3$R encourages to disentangle the semantically consistent cross-modal topics (containing some common semantic information across different modalities). In other words, the semantically co-occurring cross-modal topics are reinforced by M$^3$R through adaptively passing the mutually reinforced messages to each other in the model-learning process. To further enhance the discriminative power of the learned latent topic representations, M$^3$R incorporates the auxiliary information (i.e., categories or labels) into the process of Bayesian modeling, which boosts the modeling capability of capturing the inter-class discriminative information. Experimental results over two benchmark datasets demonstrate the effectiveness of the proposed M$^3$R in cross-modal retrieval.

Learning latent structure for activity recognition

Abstract: We present a novel latent discriminative model for human activity recognition. Unlike the approaches that require conditional independence assumptions, our model is very flexible in encoding the full connectivity among observations, latent states, and activity states. The model is able to capture richer class of contextual information in both state-state and observation-state pairs. Although loops are present in the model, we can consider the graphical model as a linear-chain structure, where the exact inference is tractable. Thereby the model is very efficient in both inference and learning. The parameters of the graphical model are learned with the Structured-Support Vector Machine (Structured-SVM). A data-driven approach is used to initialize the latent variables, thereby no hand labeling for the latent states is required. Experimental results on the CAD-120 benchmark dataset show that our model outperforms the state-of-the-art approach by over 5% in both precision and recall, while our model is more efficient in computation.

Effective ranking and search techniques for Web resources considering semantic relationships

Abstract: On the Semantic Web, the types of resources and the semantic relationships between resources are defined in an ontology. By using that information, the accuracy of information retrieval can be improved. In this paper, we present effective ranking and search techniques considering the semantic relationships in an ontology. Our technique retrieves top-k resources which are the most relevant to query keywords through the semantic relationships. To do this, we propose a weighting measure for the semantic relationship. Based on this measure, we propose a novel ranking method which considers the number of meaningful semantic relationships between a resource and keywords as well as the coverage and discriminating power of keywords. In order to improve the efficiency of the search, we prune the unnecessary search space using the length and weight thresholds of the semantic relationship path. In addition, we exploit Threshold Algorithm based on an extended inverted index to answer top-k results efficiently. The experimental results using real data sets demonstrate that our retrieval method using the semantic information generates accurate results efficiently compared to the traditional methods.

Longitudinal analysis of self-reported health status by mixture latent auto-regressive models

Abstract: Summary Motivated by an application to a longitudinal data set coming from the Health and Retirement Study about self-reported health status, we propose a model for longitudinal data which is based on a latent process to account for the unobserved heterogeneity between sample units in a dynamic fashion. The latent process is modelled by a mixture of auto-regressive AR(1) processes with different means and correlation coefficients, but with equal variances. We show how to perform maximum likelihood estimation of the proposed model by the joint use of an expectation–maximization algorithm and a Newton–Raphson algorithm, implemented by means of recursions developed in the hidden Markov model literature. We also introduce a simple method to obtain standard errors for the parameter estimates and suggest a strategy to choose the number of mixture components. In the application the response variable is ordinal; however, the approach may also be applied in other settings. Moreover, the application to the self-reported health status data set allows us to show that the model proposed is more flexible than other models for longitudinal data based on a continuous latent process. The model also achieves a goodness of fit that is similar to that of models based on a discrete latent process following a Markov chain, while retaining a reduced number of parameters. The effect of different formulations of the latent structure of the model is evaluated in terms of estimates of the regression parameters for the covariates.

Functional Principal Component Analysis of Spatiotemporal Point Processes With Applications in Disease Surveillance

Abstract: In disease surveillance applications, the disease events are modeled by spatio-temporal point processes. We propose a new class of semiparametric generalized linear mixed model for such data, where the event rate is related to some known risk factors and some unknown latent random effects. We model the latent spatio-temporal process as spatially correlated functional data, and propose Poisson maximum likelihood and composite likelihood methods based on spline approximations to estimate the mean and covariance functions of the latent process. By performing functional principal component analysis to the latent process, we can better understand the correlation structure in the point process. We also propose an empirical Bayes method to predict the latent spatial random effects, which can help highlight hot areas with unusually high event rates. Under an increasing domain and increasing knots asymptotic framework, we establish the asymptotic distribution for the parametric components in the model and the asymptotic convergence rates for the functional principal component estimators. We illustrate the methodology through a simulation study and an application to the Connecticut Tumor Registry data.

Multilevel Latent Class Analysis: Parametric and Nonparametric Models

Abstract: Latent class analysis is an analytic technique often used in educational and psychological research to identify meaningful groups of individuals within a larger heterogeneous population based on a set of variables. This technique is flexible, encompassing not only a static set of variables but also longitudinal data in the form of growth mixture modeling, as well as the application to complex multilevel sampling designs. The goal of this study was to investigate—through a Monte Carlo simulation study—the performance of several methods for parameterizing multilevel latent class analysis. Of particular interest was the comparison of several such models to adequately fit Level 1 (individual) data, given a correct specification of the number of latent classes at both levels (Level 1 and Level 2). Results include the parameter estimation accuracy as well as the quality of classification at Level 1.

Latent Fingerprint Matching: Performance Gain via Feedback from Exemplar Prints

Abstract: Latent fingerprints serve as an important source of forensic evidence in a court of law. Automatic matching of latent fingerprints to rolled/plain (exemplar) fingerprints with high accuracy is quite vital for such applications. However, latent impressions are typically of poor quality with complex background noise which makes feature extraction and matching of latents a significantly challenging problem. We propose incorporating top-down information or feedback from an exemplar to refine the features extracted from a latent for improving latent matching accuracy. The refined latent features (e.g. ridge orientation and frequency), after feedback, are used to re-match the latent to the top $K$ candidate exemplars returned by the baseline matcher and resort the candidate list. The contributions of this research include: (i) devising systemic ways to use information in exemplars for latent feature refinement, (ii) developing a feedback paradigm which can be wrapped around any latent matcher for improving its matching performance, and (iii) determining when feedback is actually necessary to improve latent matching accuracy. Experimental results show that integrating the proposed feedback paradigm with a state-of-the-art latent matcher improves its identification accuracy by 0.5-3.5 percent for NIST SD27 and WVU latent databases against a background database of 100k exemplars.

Semantic Computing of Moods Based on Tags in Social Media of Music

Abstract: Social tags inherent in online music services such as Last.fm provide a rich source of information on musical moods. The abundance of social tags makes this data highly beneficial for developing techniques to manage and retrieve mood information, and enables study of the relationships between music content and mood representations with data substantially larger than that available for conventional emotion research. However, no systematic assessment has been done on the accuracy of social tags and derived semantic models at capturing mood information in music. We propose a novel technique called Affective Circumplex Transformation (ACT) for representing the moods of music tracks in an interpretable and robust fashion based on semantic computing of social tags and research in emotion modeling. We validate the technique by predicting listener ratings of moods in music tracks, and compare the results to prediction with the Vector Space Model (VSM), Singular Value Decomposition (SVD), Nonnegative Matrix Factorization (NMF), and Probabilistic Latent Semantic Analysis (PLSA). The results show that ACT consistently outperforms the baseline techniques, and its performance is robust against a low number of track-level mood tags. The results give validity and analytical insights for harnessing millions of music tracks and associated mood data available through social tags in application development.

Latent Semantic Analysis Models on Wikipedia and TASA

Abstract: This paper introduces a collection of freely available Latent Semantic Analysis models built on the entire English Wikipedia and the TASA corpus. The models differ not only on their source, Wikipedia versus TASA, but also on the linguistic items they focus on: all words, content-words, nouns-verbs, and main concepts. Generating such models from large datasets (e.g. Wikipedia), that can provide a large coverage for the actual vocabulary in use, is computationally challenging, which is the reason why large LSA models are rarely available. Our experiments show that for the task of word-to-word similarity, the scores assigned by these models are strongly correlated with human judgment, outperforming many other frequently used measures, and comparable to the state of the art.

A Cross-Modal Approach for Extracting Semantic Relationships Between Concepts Using Tagged Images

Abstract: This paper presents a cross-modal approach for extracting semantic relationships between concepts using tagged images. In the proposed method, we first project both text and visual features of the tagged images to a latent space using canonical correlation analysis (CCA). Then, under the probabilistic interpretation of CCA, we calculate a representative distribution of the latent variables for each concept. Based on the representative distributions of the concepts, we derive two types of measures: the semantic relatedness between the concepts and the abstraction level of each concept. Because these measures are derived from a cross-modal scheme that enables the collaborative use of both text and visual features, the semantic relationships can successfully reflect semantic and visual contexts. Experiments conducted on tagged images collected from Flickr show that our measures are more coherent to human cognition than the conventional measures that use either text or visual features, or the WordNet-based measures. In particular, a new measure of semantic relatedness, which satisfies the triangle inequality, obtains the best results among different distance measures in our framework. The applicability of our measures to multimedia-related tasks such as concept clustering, image annotation and tag recommendation is also shown in the experiments.

Probabilistic Models of Cross-Lingual Semantic Similarity in Context Based on Latent Cross-Lingual Concepts Induced from Comparable Data

Abstract: We propose the first probabilistic approach to modeling cross-lingual semantic similarity (CLSS) in context which requires only comparable data. The approach relies on an idea of projecting words and sets of words into a shared latent semantic space spanned by language-pair independent latent semantic concepts (e.g., crosslingual topics obtained by a multilingual topic model). These latent cross-lingual concepts are induced from a comparable corpus without any additional lexical resources. Word meaning is represented as a probability distribution over the latent concepts, and a change in meaning is represented as a change in the distribution over these latent concepts. We present new models that modulate the isolated out-ofcontext word representations with contextual knowledge. Results on the task of suggesting word translations in context for 3 language pairs reveal the utility of the proposed contextualized models of crosslingual semantic similarity.

Semantic significance: a new measure of feature salience

Abstract: According to the feature-based model of semantic memory, concepts are described by a set of semantic features that contribute, with different weights, to the meaning of a concept. Interestingly, this theoretical framework has introduced numerous dimensions to describe semantic features. Recently, we proposed a new parameter to measure the importance of a semantic feature for the conceptual representation—that is, semantic significance. Here, with speeded verification tasks, we tested the predictive value of our index and investigated the relative roles of conceptual and featural dimensions on the participants’ performance. The results showed that semantic significance is a good predictor of participants’ verification latencies and suggested that it efficiently captures the salience of a feature for the computation of the meaning of a given concept. Therefore, we suggest that semantic significance can be considered an effective index of the importance of a feature in a given conceptual representation. Moreover, we propose that it may have straightforward implications for feature-based models of semantic memory, as an important additional factor for understanding conceptual representation.