Bayesian Weight Research Articles

An integrated approach to Bayesian weight regulations and multitasking learning methods for generating emotion-based content in the metaverse

This paper introduces an integrated model designed to analyze topics and sentiments in textual data simultaneously, recognizing their interdependence. By tackling challenges such as data scarcity, missing information, and biased distributions, the proposed model effectively captures the dynamic interactions between topics and sentiments within complex datasets. The method combines the use of Convolutional Neural Networks (CNN) for detecting topic-related patterns with a revised Recurrent Neural Network (RNN) to trace the emotional flow within contexts, leveraging the strengths of sequential data processing. These components are integrated within a Bayesian probability framework, modeling the conditional probabilities of sentences expressing specific sentiments and documents being associated with particular topics. The combined feature and state vectors from the CNN and revised RNN within this Bayesian setup enable precise classification and prediction of topics and sentiments. Furthermore, this paper explores innovative research avenues, including sentiment analysis in 3D virtual reality environments and the development of new algorithms that reflect content creation techniques in the metaverse, offering dynamic system construction. This integrated approach not only enhances data-driven decision-making but also unlocks new possibilities for advanced multi-text analysis.

Brain tumor image segmentation algorithm based on multimodal feature fusion of Bayesian weight distribution

AbstractThis study proposes an improved U‐Net model to address the issues of large semantic differences in skip connections and insufficient utilization of cross‐channel information in magnetic resonance imaging (MRI) images leading to inaccurate segmentation of brain tumor regions in the field of brain tumor segmentation. Firstly, by adding a deep residual module to alter the receptive field, the network's ability to learn tumor information is enhanced. Secondly, a dual attention mechanism was established using Bayesian weighting technology, achieving multi‐channel and multi‐scale feature fusion, and improving the model's learning and extraction of brain tumor boundary information. Finally, the tumor features extracted from different patterns are concatenated through skip connections, effectively integrating feature information from different levels and scales, and reducing semantic differences. We evaluated the performance of the proposed model on the BraTS2018 and BraTS2019 brain tumor image segmentation datasets. The experimental results showed that for the BraTS2018 dataset, the model improved the average dice score by 12.8%, the average sensitivity by 10.4%, and the average Hausdorff Distance by 5.75 compared to traditional U‐Net. On the BraTS2019 dataset, three indicators improved by 12.6%, 11.2%, and 7.46, respectively. The experimental results show that the proposed improved U‐Net model can improve the segmentation performance of brain tumor MRI images without increasing computational time.

Dataset Weighting Features Using Gain Ratio To Improve Method Accuracy Naïve Bayesian Classification

The Naïve Bayes method is proven to have a high speed when applied to large datasets, but the Naïve Bayes method has weaknesses when selecting attributes because Naïve Bayes is a statistical classification method that is only based on the Bayes theorem so that it can only be used to predict the probability of the class membership of a class independently. Independent without being able to do the selection of attributes that have a high correlation and correlation between one attribute with other attributes so that it can affect the value of accuracy. Naïve Bayesian Weight has been able to provide better accuracy than conventional Naïve Bayesian. Where an increase in the highest accuracy value obtained from the Water Quality dataset is equal to 88.57% in the Weight Naïve Bayesian classification model, while the lowest accuracy value is obtained from the Haberman dataset which is 78.95% in the conventional Naïve Bayesian classification model. The increase in accuracy of the Weight Naïve Bayesian classification model in the Water Quality dataset is 2.9%. While the increase in accuracy value in the Haberman dataset is 1.8%. If done the average accuracy of each dataset using the Weight Naïve Bayesian classification model is 2.35%. Based on the testing that has been done on all test data, it can be said that the Weight Naïve Bayesian classification model can provide better accuracy values than those produced by the conventional Naïve Bayesian classification model.

National surveillance of stroke quality of care and outcomes by applying post-stratification survey weights on the Get With The Guidelines-Stroke\xa0patient registry

BackgroundThe U.S. lacks a stroke surveillance system. This study develops a method to transform an existing registry into a nationally representative database to evaluate acute ischemic stroke care quality.MethodsTwo statistical approaches are used to develop post-stratification weights for the Get With The Guidelines-Stroke registry by anchoring population estimates to the National Inpatient Sample. Post-stratification survey weights are estimated using a raking procedure and Bayesian interpolation methods. Weighting methods are adjusted to limit the dispersion of weights and make reasonable epidemiologic estimates of patient characteristics, quality of hospital care, and clinical outcomes. Standardized differences in national estimates are reported between the two post-stratification methods for anchored and non-anchored patient characteristics to evaluate estimation quality. Primary measures evaluated are patient and hospital characteristics, stroke severity, vital and laboratory measures, disposition, and clinical outcomes at discharge.ResultsA total of 1,388,296 acute ischemic strokes occurred between 2012 and 2014. Raking and Bayesian estimates of clinical data not available in administrative data are estimated within 5 to 10% of margin for expected values. Median weight for the raking method is 1.386 and the weights at the 99th percentile is 6.881 with a maximum weight of 30.775. Median Bayesian weight is 1.329 and the 99th percentile weights is 11.201 with a maximum weight of 515.689.ConclusionsLeveraging existing databases with patient registries to develop post-stratification weights is a reliable approach to estimate acute ischemic stroke epidemiology and monitoring for stroke quality of care nationally. These methods may be applied to other diseases or settings to better monitor population health.

Extinction and discrimination in a Bayesian model of context fear conditioning (BaconX).

The extinction of contextual fear is commonly an essential requirement for successful exposure therapy for fear disorders. However, experimental work on extinction of contextual fear is limited, and there little or no directly relevant theoretical work. Here, we extend BACON, a neurocomputational model of context fear conditioning that provides plausible explanations for a number of aspects of context fear conditioning, to deal with extinction (calling the model BaconX). In this model, contextual representations are formed in the hippocampus and association of fear to them occurs in the amygdala. Representation creation, conditionability, and development of between‐session extinction are controlled by degree of confidence (assessed by the Bayesian weight of evidence) that an active contextual representation is in fact that of the current context (i.e., is “valid”). The model predicts that: (1) extinction which persists between sessions will occur only if at a sessions end there is high confidence that the active representation is valid. It follows that the shorter the context placement‐to‐US (shock) interval (“PSI”) and the less is therefore learned about context, the longer extinction sessions must be for enduring extinction to occur, while too short PSIs will preclude successful extinction. (2) Short‐PSI deficits can be rescued by contextual exposure even after conditioning has occurred. (3) Learning to discriminate well between a conditioned and similar safe context requires representations of each to form, which may not occur if PSI was too short. (4) Extinction‐causing inhibition must be applied downstream of the conditioning locus for reasonable generalization properties to be generated. (5) Context change tends to cause return of extinguished contextual fear. (6). Extinction carried out in the conditioning context generalizes better than extinction executed in contexts to which fear has generalized (as done in exposure therapy). (7) BaconX suggests novel approaches to exposure therapy.

Comprehensive evaluation model of wind power accommodation ability based on macroscopic and microscopic indicators

This article explores and establishes comprehensive evaluation index system of wind power accommodation ability considering microscopic index and macroscopic index, and the index system includes conventional evaluation indexes such as forecast deviation, simultaneity factor and anti-peak rate, also newly introduced evaluation indexes such as installed capacity, power adequacy and accommodation space. Bayesian weight modified method is used for solving index weights of 8 wind power accommodation indexes. The paper puts forward a comprehensive evaluation model of wind power accommodation ability based on improved radar chart method, and this model changes traditional radar chart fan-shaped sector to quadrilateral evaluation region, and increasing angle bisector can solve the problem that evaluation results are not unique. It constructs new area and perimeter vectors of radar chart, which make the evaluation results give consideration to level of aggregation and balance degree of evaluation objectives, and case study results show that this model has a certain practical value.

Bayesian Weight Decay on Bounded Approximation for Deep Convolutional Neural Networks.

This paper determines the weight decay parameter value of a deep convolutional neural network (CNN) that yields a good generalization. To obtain such a CNN in practice, numerical trials with different weight decay values are needed. However, the larger the CNN architecture is, the higher is the computational cost of the trials. To address this problem, this paper formulates an analytical solution for the decay parameter through a proposed objective function in conjunction with Bayesian probability distributions. For computational efficiency, a novel method to approximate this solution is suggested. This method uses a small amount of information in the Hessian matrix. Theoretically, the approximate solution is guaranteed by a provable bound and is obtained by a proposed algorithm, where its time complexity is linear in terms of both the depth and width of the CNN. The bound provides a consistent result for the proposed learning scheme. By reducing the computational cost of determining the decay value, the approximation allows for the fast investigation of a deep CNN (DCNN) which yields a small generalization error. Experimental results show that our assumption verified with different DCNNs is suitable for real-world image data sets. In addition, the proposed method significantly reduces the time cost of learning with setting the weight decay parameter while achieving good classification performances.

Joint Bayesian weight and height postnatal growth model to study the effects of maternal smoking during pregnancy.

Growth models used for describing the dynamics of body weight and height generally consider each trait independently. We proposed modeling height and weight trajectories jointly with a nonlinear heteroscedastic mixed model based on the Jenss-Bayley growth function with correlated individual random effects and using Bayesian inference techniques. Simulations showed that our model provides good estimates of the growth parameters. We illustrated how it can be used to assess the associations between maternal smoking during pregnancy, an early-life factor potentially involved in prenatal programming of obesity, and children's growth from birth to 5years of age. We used real data from the EDEN study, a large French mother-child cohort study with a high number of height and weight measurements (a total of approximately 30000 measurements for each of the 2 traits across the 1666 children). Our results supported the existence of a relationship between maternal smoking during pregnancy and growth from birth to 5years of age. Children from mothers who smoked throughout pregnancy were shown to display a higher body mass index from the first few months of life onwards compared to children from nonsmokers. At 5years of age, their mean body mass index was 0.21kg/m2 higher than unexposed children. It was mainly explained by the fact that these children tended to be smaller at birth but rapidly exceeded the weight of children from nonsmokers postnatally.

A Bayesian context fear learning algorithm/automaton.

Contextual fear conditioning is thought to involve the synaptic plasticity-dependent establishment in hippocampus of representations of to-be-conditioned contexts which can then become associated with USs in the amygdala. A conceptual and computational model of this process is proposed in which contextual attributes are assumed to be sampled serially and randomly during contextual exposures. Given this assumption, moment-to-moment information about such attributes will often be quite different from one exposure to another and, in particular, between exposures during which representations are created, exposures during which conditioning occurs, and during recall sessions. This presents challenges to current conceptual models of hippocampal function. In order to meet these challenges, our model's hippocampus was made to operate in different modes during representation creation and recall, and non-hippocampal machinery was constructed that controlled these hippocampal modes. This machinery uses a comparison between contextual information currently observed and information associated with existing hippocampal representations of familiar contexts to compute the Bayesian Weight of Evidence that the current context is (or is not) a known one, and it uses this value to assess the appropriateness of creation or recall modes. The model predicts a number of known phenomena such as the immediate shock deficit, spurious fear conditioning to contexts that are absent but similar to actually present ones, and modulation of conditioning by pre-familiarization with contexts. It also predicts a number of as yet unknown phenomena.

A center sliding Bayesian binary classifier adopting orthogonal polynomials

A center sliding Bayesian design adopting orthogonal polynomials for binary pattern classification is studied in this paper. Essentially, a Bayesian weight solution is coupled with a center sliding scheme in feature space which provides an easy tuning capability for binary classification. The proposed method is compared with several state-of-the-art binary classifiers in terms of their solution forms, decision thresholds and decision boundaries. Based on the center sliding Bayesian framework, a novel orthogonal polynomial classifier is subsequently developed. The orthogonal polynomial classifier is evaluated using two representative orthogonal polynomials for feature mapping. Our experimental results show promising potential of the orthogonal polynomial classifier since it achieves both desired accuracy and computational efficiency.

Neural networks deterioration models for serviceability condition of buried stormwater pipes

Buried stormwater pipe networks play a key role in surface drainage systems for urban areas of Australia. The pipe networks are designed to convey water from rainfall and surface runoff only and do not transport sewage. The deterioration of stormwater pipes is commonly graded into structural and serviceability condition using CCTV inspection data in order to recognize two different deterioration processes and consequences. This study investigated the application of neural networks modelling (NNM) in predicting serviceability deterioration that is associated with reductions of pipe diameter until a complete blockage. The outcomes of the NNM are predictive serviceability condition for individual pipes, which is essential for planning proactive maintenance programs, and ranking of pipe factors that potentially contribute to the serviceability deterioration. In this study the Bayesian weight estimation using Markov Chain Monte Carlo simulation was used for calibrating the NNM on a case study in order to account for the uncertainty often encountered in NNM's calibration using conventional back-propagation weight estimation. The performance and the ranked factors obtained from the NNM were also compared against a classical model using multiple discrimination analysis (MDA). The results showed that the predictive performance of the NNM using Bayesian weight estimation is better than that of the NNM using conventional backpropagation and MDA model. Furthermore, among nine input factors, ‘pipe age’ and ‘location’ appeared insignificant whilst ‘pipe size’, ‘slope’, ‘the number of trees’ and ‘climatic condition’ were found consistently important over both models for serviceability deterioration process. The remaining three factors namely, ‘structure’, ‘soil’ and ‘buried depth’ might be redundant factors. A better and more consistent data collection regime may help to improve the predictive performance of the NNM and identify the significant factors.

In‐Process Classification Assessment of Remotely Sensed Imagery

In‐Process Classification Assessment (IPCA) is concerned with the analysis and resolution of uncertainties in the classification of remotely sensed imagery. Using a Bayesian weight of evidence procedure known as Dempster‐Shafer theory, a method is presented for the mapping of three types of uncertainty: ignorance (the presence of unknown classes), ambiguity arising from inseparable training statistics, and the presence of impure (mixed) pixels. The procedure yields new images of Unknownness, Inseparability and Mixedness that are combined into a color composite for visual analysis. The color composite allows the analyst to identify classes that need improvements in training site quality and locations for new training sites to incorporate missing or mixed classes. In addition, the individual uncertainty images provide a quantitative assessment of classification quality before a post‐classification ground truth assessment is undertaken. A case study and validation show that this IPCA procedure produces an accurate reflection of classification quality.

Spatial analysis techniques applied to mineral prospecting: an evaluation in the Poços de Caldas plateau

Boolean Logic, Weighted Means, Fuzzy/AHP (Analytical Hierarchic Process), Fuzzy Gamma, and Bayesian Weight of Evidence (WofE) spatial analysis techniques were evaluated to address potential areas for radioactive mineral prospecting in the Pocos de Caldas Plateau, Brazil. These techniques were applied according to a prospecting model based on diagnostic criteria, such as favorable lithologies, structural features and gamma ray (total count) data. Resulting maps were organized into four classes of potentiality (null, low, middle and high potential), which express different levels of favorability for radioactive mineral occurrences in the study area. To compare the performance of the resulting scenarios, some 48 known radioactive mineral occurrences were used. Cross-tabulation technique was applied to measure the confidence grade (conditional probability) of each potential class. According to the results, the scenarios showed different performances. The Boolean Logic-based scenario proved to be limited, whereas result of the Fuzzy Gamma technique was unsatisfactory. However, very good results were obtained with the scenarios derived from the Bayesian WofE, Fuzzy/AHP, and Weighted Means techniques, indicating that spatial inference techniques can be useful guides in mineral prospecting campaigns.

Spatial analysis techniques applied to mineral prospecting: an evaluation in the Poços de Caldas plateau

Boolean Logic, Weighted Means, Fuzzy/AHP (Analytical Hierarchic Process), Fuzzy Gamma, and Bayesian Weight of Evidence (WofE) spatial analysis techniques were evaluated to address potential areas for radioactive mineral prospecting in the Poços de Caldas Plateau, Brazil. These techniques were applied according to a prospecting model based on diagnostic criteria, such as favorable lithologies, structural features and gamma ray (total count) data. Resulting maps were organized into four classes of potentiality (null, low, middle and high potential), which express different levels of favorability for radioactive mineral occurrences in the study area. To compare the performance of the resulting scenarios, some 48 known radioactive mineral occurrences were used. Cross-tabulation technique was applied to measure the confidence grade (conditional probability) of each potential class. According to the results, the scenarios showed different performances. The Boolean Logic-based scenario proved to be limited, whereas result of the Fuzzy Gamma technique was unsatisfactory. However, very good results were obtained with the scenarios derived from the Bayesian WofE, Fuzzy/AHP, and Weighted Means techniques, indicating that spatial inference techniques can be useful guides in mineral prospecting campaigns.

An Attempt to Provide Supplementary Information in Modelling: Bayesian Inference and Organisational Weight

An Attempt to Provide Supplementary Information in Modelling: Bayesian Inference and Organisational Weight

Bayesian weight estimation for adaptive arrays

Adaptive arrays are used in many underwater sound applications. Although there are algorithms available for updating the adaptive weight vector, most of them are obtained by considering the heuristics of a given application. In this work, an algorithm derived from a formal optimization procedure is presented. The algorithm is a recursive realization of a Bayes estimate for the weighting vector, and is in this sense optimal. Initialization of the algorithm, governed by prior knowledge of the problem, is also shown. Performance for selected situations is obtained by computer simulation, and the behavior of the algorithm as a function of several adjustable parameters is discussed in detail. [Sponsored by ONR Code 411SP.]