A novel latent factor model for recommender system

Matrix factorization (MF) has evolved as one of the better practice to handle sparse data in field of recommender systems. Funk singular value decomposition (SVD) is a variant of MF that exists as state-of-the-art method that enabled winning the Netflix prize competition. The method is widely used with modifications in present day research in field of recommender systems. With the potential of data points to grow at very high velocity, it is prudent to devise newer methods that can handle such data accurately as well as efficiently than Funk-SVD in the context of recommender system. In view of the growing data points, I propose a latent factor model that caters to both accuracy and efficiency by reducing the number of latent features of either users or items making it less complex than Funk-SVD, where latent features of both users and items are equal and often larger. A comprehensive empirical evaluation of accuracy on two publicly available, amazon and ml-100 k datasets reveals the comparable accuracy and lesser complexity of proposed methods than Funk-SVD.

Nonparametric Bayesian sparse factor models with application to gene expression modeling

A nonparametric Bayesian extension of Factor Analysis (FA) is proposed where observed data $\mathbf{Y}$ is modeled as a linear superposition, $\mathbf{G}$, of a potentially infinite number of hidden factors, $\mathbf{X}$. The Indian Buffet Process (IBP) is used as a prior on $\mathbf{G}$ to incorporate sparsity and to allow the number of latent features to be inferred. The model's utility for modeling gene expression data is investigated using randomly generated data sets based on a known sparse connectivity matrix for E. Coli, and on three biological data sets of increasing complexity.

Scaling the Indian Buffet Process via Submodular Maximization

Inference for latent feature models is inherently difficult as the inference space grows exponentially with the size of the input data and number of latent features. In this work, we use Kurihara & Welling (2008)'s maximization-expectation framework to perform approximate MAP inference for linear-Gaussian latent feature models with an Indian Buffet Process (IBP) prior. This formulation yields a submodular function of the features that corresponds to a lower bound on the model evidence. By adding a constant to this function, we obtain a nonnegative submodular function that can be maximized via a greedy algorithm that obtains at least a one-third approximation to the optimal solution. Our inference method scales linearly with the size of the input data, and we show the efficacy of our method on the largest datasets currently analyzed using an IBP model.

Religiosidade e mito na obra de José Carlos Mariátegui e a teologia da libertação

Pós-graduação em Ciências Sociais - FCLAR

TIME SERIES ANALYSIS AS INPUT FOR PREDICTIVE MODELING: PREDICTING CARDIAC ARREST IN A PEDIATRIC INTENSIVE CARE UNIT

The first manuscript, entitled "Time-Series Analysis as Input for Clinical Predictive Modeling: Modeling Cardiac Arrest in a Pediatric ICU" lays out the theoretical background for the project. There are several core concepts presented in this paper. First, traditional multivariate models (where each variable is represented by only one value) provide single point-in-time snapshots of patient status: they are incapable of characterizing deterioration. Since deterioration is consistently identified as a precursor to cardiac arrests, we maintain that the traditional multivariate paradigm is insufficient for predicting arrests. We identify time series analysis as a method capable of characterizing deterioration in an objective, mathematical fashion, and describe how to build a general foundation for predictive modeling using time series analysis results as latent variables. Building a solid foundation for any given modeling task involves addressing a number of issues during the design phase. These include selecting the proper candidate features on which to base the model, and selecting the most appropriate tool to measure them. We also identified several unique design issues that are introduced when time series data elements are added to the set of candidate features. One such issue is in defining the duration and resolution of time series elements required to sufficiently characterize the time series phenomena being considered as candidate features for the predictive model. Once the duration and resolution are established, there must also be explicit mathematical or statistical operations that produce the time series analysis result to be used as a latent candidate feature. In synthesizing the comprehensive framework for building a predictive model based on time series data elements, we identified at least four classes of data that can be used in the model design. The first two classes are shared with traditional multivariate models: multivariate data and clinical latent features. Multivariate data is represented by the standard one value per variable paradigm and is widely employed in a host of clinical models and tools. These are often represented by a number present in a given cell of a table. Clinical latent features derived, rather than directly measured, data elements that more accurately represent a particular clinical phenomenon than any of the directly measured data elements in isolation. The second two classes are unique to the time series data elements. The first of these is the raw data elements. These are represented by multiple values per variable, and constitute the measured observations that are typically available to end users when they review time series data. These are often represented as dots on a graph. The final class of data results from performing time series analysis. This class of data represents the fundamental concept on which our hypothesis is based. The specific statistical or mathematical operations are up to the modeler to determine, but we generally recommend that a variety of analyses be performed in order to maximize the likelihood that a representation of the time series data elements is produced that is able to distinguish between two or more classes of outcomes. The second manuscript, entitled "Building Clinical Prediction Models Using Time Series Data: Modeling Cardiac Arrest in a Pediatric ICU" provides a detailed description, start to finish, of the methods required to prepare the data, build, and validate a predictive model that uses the time series data elements determined in the first paper. One of the fundamental tenets of the second paper is that manual implementations of time series based models are unfeasible due to the relatively large number of data elements and the complexity of preprocessing that must occur before data can be presented to the model. Each of the seventeen steps is analyzed from the perspective of how it may be automated, when necessary. We identify the general objectives and available strategies of each of the steps, and we present our rationale for choosing a specific strategy for each step in the case of predicting cardiac arrest in a pediatric intensive care unit. Another issue brought to light by the second paper is that the individual steps required to use time series data for predictive modeling are more numerous and more complex than those used for modeling with traditional multivariate data. Even after complexities attributable to the design phase (addressed in our first paper) have been accounted for, the management and manipulation of the time series elements (the preprocessing steps in particular) are issues that are not present in a traditional multivariate modeling paradigm. In our methods, we present the issues that arise from the time series data elements: defining a reference time; imputing and reducing time series data in order to conform to a predefined structure that was specified during the design phase; and normalizing variable families rather than individual variable instances. The final manuscript, entitled: "Using Time-Series Analysis to Predict Cardiac Arrest in a Pediatric Intensive Care Unit" presents the results that were obtained by applying the theoretical construct and its associated methods (detailed in the first two papers) to the case of cardiac arrest prediction in a pediatric intensive care unit. Our results showed that utilizing the trend analysis from the time series data elements reduced the number of classification errors by 73%. The area under the Receiver Operating Characteristic curve increased from a baseline of 87% to 98% by including the trend analysis. In addition to the performance measures, we were also able to demonstrate that adding raw time series data elements without their associated trend analyses improved classification accuracy as compared to the baseline multivariate model, but diminished classification accuracy as compared to when just the trend analysis features were added (ie, without adding the raw time series data elements). We believe this phenomenon was largely attributable to overfitting, which is known to increase as the ratio of candidate features to class examples rises. Furthermore, although we employed several feature reduction strategies to counteract the overfitting problem, they failed to improve the performance beyond that which was achieved by exclusion of the raw time series elements. Finally, our data demonstrated that pulse oximetry and systolic blood pressure readings tend to start diminishing about 10-20 minutes before an arrest, whereas heart rates tend to diminish rapidly less than 5 minutes before an arrest.

Supervised latent dirichlet allocation models for efficient activity representation

Local spatio-temporal features with a Bag-of-visual words model is a popular approach used in human action recognition. Bag-of-features methods suffer from several challenges such as extracting appropriate appearance and motion features from videos, converting extracted features appropriate for classification and designing a suitable classification framework. In this paper we address the problem of efficiently representing the extracted features for classification to improve the overall performance. We introduce two generative supervised topic models, maximum entropy discrimination LDA (MedLDA) and class- specific simplex LDA (css-LDA), to encode the raw features suitable for discriminative SVM based classification. Unsupervised LDA models disconnect topic discovery from the classification task, hence yield poor results compared to the baseline Bag-of-words framework. On the other hand supervised LDA techniques learn the topic structure by considering the class labels and improve the recognition accuracy significantly. MedLDA maximizes likelihood and within class margins using max-margin techniques and yields a sparse highly discriminative topic structure; while in css-LDA separate class specific topics are learned instead of common set of topics across the entire dataset. In our representation first topics are learned and then each video is represented as a topic proportion vector, i.e. it can be comparable to a histogram of topics. Finally SVM classification is done on the learned topic proportion vector. We demonstrate the efficiency of the above two representation techniques through the experiments carried out in two popular datasets. Experimental results demonstrate significantly improved performance compared to the baseline Bag-of-features framework which uses kmeans to construct histogram of words from the feature vectors.

A Bayesian latent class safety performance function for identifying motor vehicle crash blackspots

The current state of the practice in Blackspot Identification (BSI) utilizes safety performance functions based on total crash counts to identify transport system sites with potentially high crash risk. This paper postulates that total crash count variation over a transport network is a result of multiple distinct crash generating processes including geometric characteristics of the road, spatial features of the surrounding environment, and driver behaviour factors. However, these multiple sources are ignored in current modelling methodologies in both trying to explain or predict crash frequencies across sites. Instead, current practice employs models that imply that a single underlying crash generating process exists. The model mis-specification may lead to correlating crashes with the incorrect sources of contributing factors (e.g. concluding a crash is predominately caused by a geometric feature when it is a behavioural issue), which may ultimately lead to inefficient use of public funds and misidentification of true blackspots. This study aims to propose a latent class model consistent with a multiple crash process theory, and to investigate the influence this model has on correctly identifying crash blackspots. We first present the theoretical and corresponding methodological approach in which a Bayesian Latent Class (BLC) model is estimated assuming that crashes arise from two distinct risk generating processes including engineering and unobserved spatial factors. The Bayesian model is used to incorporate prior information about the contribution of each underlying process to the total crash count. The methodology is applied to the state-controlled roads in Queensland, Australia and the results are compared to an Empirical Bayesian Negative Binomial (EB-NB) model. A comparison of goodness of fit measures illustrates significantly improved performance of the proposed model compared to the NB model. The detection of blackspots was also improved when compared to the EB-NB model. In addition, modelling crashes as the result of two fundamentally separate underlying processes reveals more detailed information about unobserved crash causes.

On Finding the Natural Number of Topics with Latent Dirichlet Allocation: Some Observations

It is important to identify the ``correct'' number of topics in mechanisms like Latent Dirichlet Allocation(LDA) as they determine the quality of features that are presented as features for classifiers like SVM. In this work we propose a measure to identify the correct number of topics and offer empirical evidence in its favor in terms of classification accuracy and the number of topics that are naturally present in the corpus. We show the merit of the measure by applying it on real-world as well as synthetic data sets(both text and images). In proposing this measure, we view LDA as a matrix factorization mechanism, wherein a given corpus C is split into two matrix factors M-1 and M-2 as given by C-d*w = M1(d*t) x Q(t*w).Where d is the number of documents present in the corpus anti w is the size of the vocabulary. The quality of the split depends on ``t'', the right number of topics chosen. The measure is computed in terms of symmetric KL-Divergence of salient distributions that are derived from these matrix factors. We observe that the divergence values are higher for non-optimal number of topics - this is shown by a `dip' at the right value for `t'.

Some salient features of the atmosphere observed over the north Bay of Bengal during BOBMEX

This paper describes the near surface characteristics and vertical variations based on the observations made at 17.5degreesN and 89degreesE from ORV Sagar Kanya in the north Bay of Bengal during the Bay of Bengal Monsoon Experiment (BOBMEX) carried out in July-August 1999. BOBMEX captured both the active and weak phases of convection. SST remained above the convection threshold throughout the BOBMEX. While the response of the SST to atmospheric forcing was clearly observed, the response of the atmosphere to SST changes was not clear. SST decreased during periods of large scale precipitation, and increased during a weak phase of convection. It is shown that the latent heat flux at comparable wind speeds was about 25-50% lower over the Bay during BOBMEX compared to that over the Indian Ocean during other seasons and tropical west Pacific. On the other hand, the largest variations in the surface daily net heat flux are observed over the Bay during BOBMEX. SST predicted using observed surface fluxes showed that 1-D heat balance model works sometime but not always, and horizontal advection is important. The high resolution Vaisala radiosondes launched during BOBMEX could clearly bring out the changes in the vertical structure of the atmosphere between active and weak phases of convection. Convective Available Potential Energy of the surface air decreased,by 2-3 kJ kg(-1) following convection, and recovered in a time period of one or two days. The mid tropospheric relative humidity and water vapor content, and wind direction show the major changes between the active and weak phases of convection.

CONFORMATIONAL FEATURES OF SIGMA FACTOR/ANTI-SIGMA COMPLEXES

The association of a factors with the RNA polymerase dictates the expression profile of a bacterial cell. Major changes to the transcription profile are achieved by the use of multiple sigma factors that confer distinct promoter selectivity to the holoenzyme. The cellular concentration of a sigma factor is regulated by diverse mechanisms involving transcription, translation and post-translational events. The number of sigma factors varies substantially across bacteria. The diversity in the interactions between sigma factors also vary-ranging from collaboration, competition or partial redundancy in some cellular or environmental contexts. These interactions can be rationalized by a mechanistic model referred to as the partitioning of a space model of bacterial transcription. The structural similarity between different sigma/anti-sigma complexes despite poor sequence conservation and cellular localization reveals an elegant route to incorporate diverse regulatory mechanisms within a structurally conserved scaffold. These features are described here with a focus on sigma/anti-sigma complexes from Mycobacterium tuberculosis. In particular, we discuss recent data on the conditional regulation of sigma/anti-sigma factor interactions. Specific stages of M. tuberculosis infection, such as the latent phase, as well as the remarkable adaptability of this pathogen to diverse environmental conditions can be rationalized by the synchronized action of different a factors.

A meso elastoplastic constitutive model for polycrystalline metals based on equivalent slip systems with latent hardening

A meso material model for polycrystalline metals is proposed, in which the tiny slip systems distributing randomly between crystal slices in micro-grains or on grain boundaries are replaced by macro equivalent slip systems determined by the work-conjugate principle. The elastoplastic constitutive equation of this model is formulated for the active hardening, latent hardening and Bauschinger effect to predict macro elastoplastic stress-strain responses of polycrystalline metals under complex loading conditions. The influence of the material property parameters on size and shape of the subsequent yield surfaces is numerically investigated to demonstrate the fundamental features of the proposed material model. The derived constitutive equation is proved accurate and efficient in numerical analysis. Compared with the self-consistent theories with crystal grains as their basic components, the present theory is much simpler in mathematical treatment.

Spatio-temporal Bayesian network models with latent variables for revealing trophic dynamics and functional networks in fisheries ecology

Ecosystems consist of complex dynamic interactions among species and the environment, the understanding of which has implications for predicting the environmental response to changes in climate and biodiversity. However, with the recent adoption of more explorative tools, like Bayesian networks, in predictive ecology, few assumptions can be made about the data and complex, spatially varying interactions can be recovered from collected field data. In this study, we compare Bayesian network modelling approaches accounting for latent effects to reveal species dynamics for 7 geographically and temporally varied areas within the North Sea. We also apply structure learning techniques to identify functional relationships such as prey–predator between trophic groups of species that vary across space and time. We examine if the use of a general hidden variable can reflect overall changes in the trophic dynamics of each spatial system and whether the inclusion of a specific hidden variable can model unmeasured group of species. The general hidden variable appears to capture changes in the variance of different groups of species biomass. Models that include both general and specific hidden variables resulted in identifying similarity with the underlying food web dynamics and modelling spatial unmeasured effect. We predict the biomass of the trophic groups and find that predictive accuracy varies with the models' features and across the different spatial areas thus proposing a model that allows for spatial autocorrelation and two hidden variables. Our proposed model was able to produce novel insights on this ecosystem's dynamics and ecological interactions mainly because we account for the heterogeneous nature of the driving factors within each area and their changes over time. Our findings demonstrate that accounting for additional sources of variation, by combining structure learning from data and experts' knowledge in the model architecture, has the potential for gaining deeper insights into the structure and stability of ecosystems. Finally, we were able to discover meaningful functional networks that were spatially and temporally differentiated with the particular mechanisms varying from trophic associations through interactions with climate and commercial fisheries.

A latent feature analysis of the neural representation of conceptual knowledge

Bayesian probabilistic analysis offers a new approach to characterize semantic representations by inferring the most likely feature structure directly from the patterns of brain activity. In this study, infinite latent feature models [1] are used to recover the semantic features that give rise to the brain activation vectors when people think about properties associated with 60 concrete concepts. The semantic features recovered by ILFM are consistent with the human ratings of the shelter, manipulation, and eating factors that were recovered by a previous factor analysis. Furthermore, different areas of the brain encode different perceptual and conceptual features. This neurally-inspired semantic representation is consistent with some existing conjectures regarding the role of different brain areas in processing different semantic and perceptual properties. © 2012 Springer-Verlag.