Bayesian Inference Methods Research Articles

Integrated data analysis on the electron density profile of HL-2A with the Bayesian probability inference method

In fusion research, the diagnostic data are obtained from different diagnostic systems, which are relatively independent (in terms of the response function, noise, calibration, etc…). The consequence is that many measurements providing the same physical quantity could provide different results. In this work, the Bayesian probability inference has been applied to the frequency modulated continuous wave reflectometry and the Far-infrared laser interferometer diagnostic systems on HL-2A tokamak, offering the integrated data analysis (IDA) for electron density profile reconstruction. With the implementation, it is demonstrated that more comprehensive inference could be delivered from IDA compared to the traditional individual data analysis technique. The data analysis program based on the Bayesian inference model has been developed to reconstruct the two-dimensional electron density profile, which permits to be further implementation of the HL-2A/2M IDA framework in the near future.

Description of the Three Complete Mitochondrial Genomes of Click Beetles (Coleoptera, Elateridae) with Phylogenetic Implications

The family Elateridae, known as click beetles, is a mega-diverse lineage of Coleoptera. Wireworms are the larval stage of click beetles, which are generalist herbivores and which are recognized as economically important pests of crops. To more effectively control and monitor wireworms, it is crucial to understand the genetics, taxonomy and phylogenetics of Elateridae. Here, we sequenced and characterized three complete mitochondrial genomes (mitogenomes) from the subfamily Elaterinae using a next-generation sequencing approach. In addition, we provided the annotated mitogenomes of the newly sequenced species, namely Parasilesis musculus (Candèze, 1873), Melanotus cribricollis Candèze, 1860 and Glyphonyx sp., and compared their arrangement with other closely related species. The secondary structures of tRNA genes and rRNA genes were predicted. Combined with the published mitogenomes of elaterid species, we reconstructed the phylogenetic framework for Elateridae under maximum likelihood and Bayesian inference methods using nucleotide and amino acid sequence datasets separately. The results from the Bayesian analysis based on the nucleotide dataset PCGRNA including all 37 mitochondrial genes were congruent with previous studies. Within the monophyletic Elateridae, two main clades were recovered. The first clade included Elaterinae and Melanotus. The second clade consisted of the remaining subfamilies. Physodactylinae and Cardiophorinae formed a sister group. Agrypninae was monophyletic. A subclade comprised Negastriinae and Dendrometrinae.

Current and future trends in heat-related mortality in the MENA region: a health impact assessment with bias-adjusted statistically downscaled CMIP6 (SSP-based) data and Bayesian inference

The Middle East and North Africa (MENA) is one of the regions that is most vulnerable to the negative effects of climate change, yet the potential public health impacts have been underexplored compared to other regions. We aimed to examine one aspect of these impacts, heat-related mortality, by quantifying the current and future burden in the MENA region and identifying the most vulnerable countries. We did a health impact assessment using an ensemble of bias-adjusted statistically downscaled Coupled Model Intercomparison Project phase 6 (CMIP6) data based on four Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2·6 [consistent with a 2°C global warming scenario], SSP2-4·5 [medium pathway scenario], SSP3-7·0 [pessimistic scenario], and SSP5-8·5 [high emissions scenario]) and Bayesian inference methods. Assessments were based on apparent temperature-mortality relationships specific to each climate subregion of MENA based on Koppen-Geiger climate type classification, and unique thresholds were characterised for each 50 km grid cell in the region. Future annual heat-related mortality was estimated for the period 2021-2100. Estimates were also presented with population held constant to quantify the contribution of projected demographic changes to the future heat-mortality burden. The average annual heat-related death rate across all MENA countries is currently 2·1 per 100 000 people. Under the two high emissions scenarios (SSP3-7·0 and SSP5-8·5), most of the MENA region will have experienced substantial warming by the 2060s. Annual heat-related deaths of 123·4 per 100 000 people are projected for MENA by 2100 under a high emissions scenario (SSP5-8·5), although this rate would be reduced by more than 80% (to 20·3 heat-related deaths per 100 000 people per year) if global warming could be limited to 2°C (ie, under the SSP1-2·6 scenario). Large increases are also expected by 2100 under the SSP3-7·0 scenario (89·8 heat-related deaths per 100 000 people per year) due to the high population growth projected under this pathway. Projections in MENA are far higher than previously observed in other regions, with Iran expected to be the most vulnerable country. Stronger climate change mitigation and adaptation policies are needed to avoid these heat-related mortality impacts. Since much of this increase will be driven by population changes, demographic policies and healthy ageing will also be key to successful adaptation. National Institute for Health Research, EU Horizon 2020.

The first two mitochondrial genomes from Apiotrichum reveal mitochondrial evolution and different taxonomic assignment of Trichosporonales

Apiotrichum is a diverse anamorphic basidiomycetous yeast genus, and its mitogenome characterization has not been revealed. In this study, we assembled two Apiotrichum mitogenomes and compared them with mitogenomes from Agaricomycotina, Pucciniomycotina and Ustilaginomycotina. The mitogenomes of Apiotrichum gracile and A. gamsii comprised circular DNA molecules, with sizes of 34,648 bp and 38,096 bp, respectively. Intronic regions were found contributed the most to the size expansion of A. gamsii mitogenome. Comparative mitogenomic analysis revealed that 6.85–38.89% of nucleotides varied between tRNAs shared by the two Apiotrichum mitogenomes. The GC content of all core PCGs in A. gamsii was lower than that of A. gracile, with an average low value of 4.97%. The rps3 gene differentiated the most among Agaricomycotina, Pucciniomycotina and Ustilaginomycotina species, while nad4L gene was the most conserved in evolution. The Ka/Ks values for cob and rps3 genes were > 1, indicating the two genes may be subjected to positive selection in Agaricomycotina, Pucciniomycotina and Ustilaginomycotina. Frequent intron loss/gain events and potential intron transfer events have been detected in evolution of Agaricomycotina, Pucciniomycotina and Ustilaginomycotina. We further detected large-scale gene rearrangements between the 19 mitogenomes from Agaricomycotina, Pucciniomycotina and Ustilaginomycotina, and fifteen of the 17 mitochondrial genes shared by Apiotrichum varied in gene arrangements. Phylogenetic analyses based on maximum likelihood and Bayesian inference methods using a combined mitochondrial gene dataset revealed different taxonomic assignment of two Apiotrichum species, wherein A. gamsii had a more closely relationship with Trichosporon asahii. This study served as the first report on mitogenomes from the genus Apiotrichum, which promotes the understanding of evolution, genomics, and phylogeny of Apiotrichum.

BI-IEnKF coupling model for effective source term estimation of natural gas leakage in urban utility tunnels

As an effective way to facilitate the increasing demand for reliable infrastructure, energy supply and sustainable urban development, underground utility tunnels have been developed rapidly in recent years. Due to the widespread distribution of utility tunnels, the safe operation of natural gas pipelines accommodated in utility tunnels has caused great concern considering fire, explosion, and other coupling consequences induced by the gas pipeline leakage. However, the limited information on leakage source terms in accidental leakage scenarios could preclude timely consequence assessment and effective emergency response. In this study, a BI-IEnKF coupling source term estimation (STE) model is developed, with the combination of gas dispersion model, Bayesian inference (BI) and iterative ensemble Kalman filter (IEnKF) method, to achieve the effective source term estimation (including leakage location and leakage rate) and gas concentration distribution prediction. The newly developed model is first evaluated by the twin experiment with good reliability and accuracy. Furthermore, three contributing factors affecting the performance of the developed BI-IEnKF coupling STE model were investigated to assist parameter selection for practical use. Additionally, the novel application of mobile sensors serving as an alternative for fixed sensors is explored, and an application framework is sequentially given to guide the deployment of the developed coupling model in utility tunnels. The results show that the developed model has great performance in accuracy, efficiency and robustness, as well as the potential to be applied in actual utility tunnel scenarios. This study can provide technical supports for safety control and emergency response in the case of natural gas pipeline leakage accidents in utility tunnels. Also, it could be helpful to reasonable references for gas lekage monitoring system design.

Efficient exploration of reionization parameters for the upcoming 21 cm observations using a photon-conserving seminumerical model SCRIPT

ABSTRACTOne of the most promising probes to constrain the reionization history of the universe is the power spectrum of neutral hydrogen 21 cm emission fluctuations. The corresponding analyses require computationally efficient modelling of reionization, usually achieved through seminumerical simulations. We investigate the capability of one such seminumerical code, SCRIPT, to constrain the reionization parameters. Our study involves creating a mock data set corresponding to the upcoming SKA-Low, followed by a Bayesian inference method to constrain the model parameters. In particular, we explore in detail whether the inferred parameters are unbiased with respect to the inputs used for the mock and also if the inferences are insensitive to the resolution of the simulation. We find that the model is successful on both fronts. We also develop a simple template model of reionization that can mimic the complex physical processes like inhomogeneous recombinations and radiative feedback and show that it can recover the global reionization history reliably with a moderate computational cost. However, such simple models are not suitable for constraining the properties of the ionizing sources. Our results are relevant for constraining reionization using high-quality data expected in the future telescopes.

Five new species of Schizoporaceae (Basidiomycota, Hymenochaetales) from East Asia.

Five new wood-inhabiting fungi, Lyomycesalbopulverulentus, L.yunnanensis, Xylodondaweishanensis, X.fissuratus, and X.puerensisspp. nov., are proposed based on a combination of morphological features and molecular evidence. Lyomycesalbopulverulentus is characterized by brittle basidiomata, pruinose hymenophore with a white hymenial surface, a monomitic hyphal system with clamped generative hyphae, and ellipsoid basidiospores. Lyomycesyunnanensis is characterized by a grandinioid hymenial surface, the presence of capitate cystidia, and ellipsoid basidiospores. Xylodondaweishanensis is characterized by an odontioid hymenial surface, a monomitic hyphal system with clamped generative hyphae, and broad ellipsoid-to-subglobose basidiospores. Xylodonfissuratus is characterized by a cracking basidiomata with a grandinioid hymenial surface, and ellipsoid basidiospores. Xylodonpuerensis is characterized by a poroid hymenophore with an angular or slightly daedaleoid configuration, and ellipsoid-to-broad-ellipsoid basidiospores. Sequences of ITS and nLSU rRNA markers of the studied samples were generated and phylogenetic analyses were performed with the maximum likelihood, maximum parsimony, and Bayesian inference methods. The phylogram based on the ITS+nLSU rDNA gene regions (Fig. 1) included six genera within the families Chaetoporellaceae, Hyphodontiaceae, Hymenochaetaceae, and Schizoporaceae (Hymenochaetales)-Fasciodontia, Hastodontia, Hyphodontia, Kneifiella, Lyomyces, and Xylodon-in which the five new species were grouped into genera Lyomyces and Xylodon. The phylogenetic tree inferred from the ITS sequences highlighted that Lyomycesalbopulverulentus formed a monophyletic lineage and was then grouped closely with L.bambusinus, L.orientalis, and L.sambuci; additionally, L.yunnanensis was sister to L.niveus with strong supports. The topology, based on the ITS sequences, revealed that Xylodondaweishanensis was retrieved as a sister to X.hyphodontinus; X.fissuratus was grouped with the four taxa X.montanus, X.subclavatus, X.wenshanensis, and X.xinpingensis; and X.puerensis was clustered with X.flaviporus, X.ovisporus, X.subflaviporus, X.subtropicus, and X.taiwanianus.

Chrysoporthe puriensis causing canker and mortality in Pleroma mutabile in the Atlantic Forest, Brazil

AbstractFungi of the Cryphonectriaceae family are globally known to be tree pathogens. In Brazil, several species of Chrysoporthe have been found causing stem and branch cankers in Pleroma (= Tibouchina) spp. Recently, Chrysoporthe puriensis was described as a new species causing stem and branch cankers in Pleroma granulosum, Pleroma candolleanum, and Pleroma heteromallum, all native species of the Melastomataceae family. During an investigation to collect isolates of the Cryphonectriaceae family in an important Brazilian biome, the Atlantic Forest, in Serra do Mar, structures typical of Chrysoporthe species were found in a different host, Pleroma mutabile. Fruiting bodies present in the bark of these trees were collected and isolated. The isolates were submitted on morphological characterization and phylogenetic analyses of the internal transcribed spacer (ITS) and β‐tubulin gene regions using Maximum likelihood, Bayesian inference and Maximum parsimony methods. The isolates collected, for this study, were identified as C. puriensis. Pathogenicity tests on seedlings of Pleroma species and Eucalyptus clones revealed C. puriensis can infect and cause canker in these plant species as mortality. The results demonstrate the importance of delimiting the C. puriensis collection range to track its dissemination in other hosts. No host specificity was observed in the inoculation tests, suggesting this is an important finding, the pathogen causes diseases and mortality in several plants of the Atlantic Forest. Additionally, the pathogen can affect others hosts, such as Eucalyptus clones in commercial plantations.

Complete mitochondrial genome sequence and annotation of Rhinogobius lentiginis (Gobiiformes: Gobiidae: Gobionellinae)

We report the complete mitochondrial genome of Rhinogobius lentiginis, which was found to be a circular molecule of 16,633 bp in length and included 13 protein-coding genes (PCGs), 2 rRNA genes, 22 tRNA genes, and a non-coding control region. The overall base composition was 28.44% A, 26.21% T, 16.33% G, and 29.02% C. Phylogenetic analyses using maximum-likelihood and Bayesian inference methods revealed a close genome relationship among R. lentiginis, R. niger, R. shennongensis and R. maculagenys. The complete mitogenome of R. lentiginis will provide a valuable resource for species classification and conservation.

Molecular Phylogenetic Relationships and Unveiling Novel Genetic Diversity among Slow and Pygmy Lorises, including Resurrection of Xanthonycticebus intermedius.

Genetic analysis of historical museum collections presents an opportunity to clarify the evolutionary history of understudied primate groups, improve taxonomic inferences, and inform conservation efforts. Among the most understudied primate groups, slow and pygmy lorises (genera Nycticebus and Xanthonycticebus) are nocturnal strepsirrhines found in South and Southeast Asia. Previous molecular studies have supported five species, but studies using morphological data suggest the existence of at least nine species. We sequenced four mitochondrial loci, CO1, cytb, d-loop, and ND4, for a total of 3324 aligned characters per sample from 41 historical museum specimens for the most comprehensive geographic coverage to date for these genera. We then combined these sequences with a larger dataset composed of samples collected in Vietnam as well as previously published sequences (total sample size N = 62). We inferred phylogenetic relationships using Bayesian inference and maximum likelihood methods based on data from each locus and on concatenated sequences. We also inferred divergence dates for the most recent common ancestors of major lineages using a BEAST analysis. Consistent with previous studies, we found support for Xanthonycticebus pygmaeus as a basal taxon to the others in the group. We also confirmed the separation between lineages of X. pygmaeus from northern Vietnam/Laos/China and southern Vietnam/Cambodia and included a taxonomic revision recognizing a second taxon of pygmy loris, X. intermedius. Our results found support for multiple reciprocally monophyletic taxa within Borneo and possibly Java. The study will help inform conservation management of these trade-targeted animals as part of a genetic reference database for determining the taxonomic unit and provenance of slow and pygmy lorises confiscated from illegal wildlife trade activities.

Specifying prior distributions in reliability applications

Especially when facing reliability data with limited information (e.g., a small number of failures), there are strong motivations for using Bayesian inference methods. These include the option to use information from physics‐of‐failure or previous experience with a failure mode in a particular material to specify an informative prior distribution. Another advantage is the ability to make statistical inferences without having to rely on specious (when the number of failures is small) asymptotic theory needed to justify non‐Bayesian methods. Users of non‐Bayesian methods are faced with multiple methods of constructing uncertainty intervals (Wald, likelihood, and various bootstrap methods) that can give substantially different answers when there is little information in the data. For Bayesian inference, there is only one method of constructing equal‐tail credible intervals—but it is necessary to provide a prior distribution to fully specify the model. Much work has been done to find default prior distributions that will provide inference methods with good (and in some cases exact) frequentist coverage properties. This paper reviews some of this work and provides, evaluates, and illustrates principled extensions and adaptations of these methods to the practical realities of reliability data (e.g., non‐trivial censoring).

New Mitogenomes of the Polypedilum Generic Complex (Diptera: Chironomidae): Characterization and Phylogenetic Implications.

Mitochondrial genomics, as a useful marker for phylogenetics and systematics of organisms, are important for molecular biology studies. The phylogenetic relationships of the Polypedilum generic complex remains controversial, due to lack taxonomy and molecular information. In this study, we newly sequenced mitogenomes of 14 species of the Polypedilum generic complex. Coupled with three recently published sequences, we analyzed the nucleotide composition, sequence length, and evolutionary rate of this generic complex. The control region showed the highest AT content. The evolution rate of protein coding genes was as follows: ATP8 > ND6 > ND5 > ND3 > ND2 > ND4L > ND4 > COX1 > ND1 > CYTB > APT6 > COX2 > COX3. We reconstructed the phylogenetic relationships among the genera within the Polypedilum generic complex based on 19 mitochondrial genomes (seventeen ingroups and two outgroups), using Bayesian Inference (BI) and Maximum Likelihood (ML) methods for all databases. Phylogenetic analysis of 19 mitochondrial genomes demonstrated that the Endochironomus + Synendotendipes was sister to Phaenopsectra + Sergentia.

Do Narcissistic Admiration and Rivalry Matter for the Mental Health of Romantic Partners? Insights from Actor-Partner Interdependence Models

Personality traits of romantic partners may be part of the puzzle of how romantic relationships are related to mental health. We investigated the role of narcissistic admiration and rivalry in this context. Positive associations of admiration and negative associations of rivalry with the mental health of individuals and their partners were hypothesized. Furthermore, we expected admiration to be particularly relevant for the narcissist’s mental health and rivalry to matter more for the partner. Actor-partner interdependence models and Bayesian inference methods were used ( n = 7438 couples). Hypotheses and methods were pre-registered. Contrary to expectations, the results only supported the hypotheses concerning actor effects. Most partner effects were miniscule, with the exception of the effect of male rivalry on female mental health. While controlling for extraversion and agreeableness reduced and partly reversed the initial effects of narcissism, an exploratory analysis suggested that these personality factors should be considered as mediators and that most partner effects of admiration and rivalry on mental health may be cases of indirect-only mediation, with extraversion and agreeableness acting as process variables. While replications are needed, the findings suggest that the question posed in the title may be answered with “not that much, apparently”.

A Bayesian Approach for Characterizing and Mitigating Gate and Measurement Errors

Various noise models have been developed in quantum computing study to describe the propagation and effect of the noise that is caused by imperfect implementation of hardware. Identifying parameters such as gate and readout error rates is critical to these models. We use a Bayesian inference approach to identify posterior distributions of these parameters such that they can be characterized more elaborately. By characterizing the device errors in this way, we can further improve the accuracy of quantum error mitigation. Experiments conducted on IBM’s quantum computing devices suggest that our approach provides better error mitigation performance than existing techniques used by the vendor. Also, our approach outperforms the standard Bayesian inference method in some scenarios.

State estimation based on enhanced Bayesian approach: Application in water distribution systems

Hydraulic models have been regarded as an essential tool for controlling and operating water distribution systems (WDSs). With the rapid development of communication networks, real-time control is becoming a hot issue. The nodal water demand as the state variable should be estimated for the control and management of WDS. Due to the lack of sufficient observed data to determine the nodal water demand, the hydraulic model may result in unreasonable model outputs, which are undesirable in actual engineering. We develop a Bayesian inference method to fuse nodal pressure prior information into the state estimation process to address this problem. The developed approach can constrain the nodal water demand estimation process and keep the model outputs (e.g., nodal water pressure) in the feasible domain. This approach is applied to a simple network and a realistic large-scale network. Results show that prior nodal pressure information can significantly increase estimation accuracy and decrease the uncertainty, and the estimated nodal pressure can be confined to a feasible domain.

Higher-level phylogeny and evolutionary history of nonditrysians (Lepidoptera) inferred from mitochondrial genome sequences

Abstract As one of the largest single-radiation sources of phytophagous insects, the higher level phylogeny of lepidopteran lineages has been debated for decades, especially the early divergences of nonditrysia. Here, we use 74 mitochondrial genomes representing seven nonditrysian superfamilies, including the newly sequenced mitochondrial genomes of eight families and 30 species. Four datasets were constructed and used for a phylogenetic analysis based on maximum likelihood (ML) and Bayesian inference (BI) methods. The results reveal that (1) nonditrysian lineages presented typically nested relationships with the family Micropterigidae as the sister-group to the remaining Lepidoptera, (2) monophyly of major clades in Lepidoptera, including Glossata, Heteroneura, Eulepidoptera and Ditrysia, was strongly supported, and (3) Neopseustoidea formed the sister of the remaining Glossata. Divergence-time analysis estimated that Lepidoptera originated c. 222 Mya during the Middle Triassic and that most superfamilies/families of nonditrysia originated from the Early Cretaceous to the Early Palaeogene, which largely coincided with the radiation of angiosperms. Additionally, three novel gene rearrangements in the gene cluster ‘MIQ’ and ‘TP’ were found in nonditrysian groups. Further mitochondrial genome data will help us to better understand the evolution of early nonditrysian lineages.

Probing the baryon mass fraction in IGM and its redshift evolution with fast radio bursts using Bayesian inference method

ABSTRACT We investigate the fraction of baryon mass in intergalactic medium (fIGM), using 18 well-localized fast radio bursts (FRBs) in the redshift range z ∈ (0.0039, 0.66). We construct a five-parameter Bayesian inference model, with the probability distributions of dispersion measures (DMs) of IGM and host galaxy properly taken into account. To check the possible redshift evolution, we parametrize fIGM as a mildly evolving function of redshift, fIGM = fIGM,0[1 + αz/(1 + z)]. By simultaneously constraining five parameters, we get $f_\mathrm{IGM,0} = 0.92^{+0.06}_{-0.12}$ and $\alpha = 0.49^{+0.59}_{-0.47}$, and the median value of DM of host galaxy is $\exp (\mu)=72.49^{+33.31}_{-25.62}~{\rm pc ~ cm ^ {-3}}$. By fixing two parameters that can be constrained independently with other observations, we obtain $\alpha =0.11^{+0.24}_{-0.27}$ in the three-parameter fit, which is consistent with zero within 1σ uncertainty. Monte Carlo simulations show that even 300 FRBs are not enough to tightly constrain five parameters simultaneously. This is mainly caused by the correlation between parameters. Only if two parameters are fixed, 100 FRBs are necessary to achieve unbiased constraints on the remaining parameters.

Divergence time of mites of the family Laelapidae based on mitochondrial barcoding region.

Using the mitochondrial barcoding region to correlate research with 58 species in 19 genera of the family Laelapidae with the aim of determining the origin, phylogenetic relationships, and biogeographic historical distribution characteristics of mites in the family Laelapidae. Phylogenetic trees were obtained using Bayesian inference (BI) and Maximum-likelihood (ML) methods, based on three fossil records calibrated as molecular clock nodes, to estimate the divergence time of mites in the family Laelapidae as well as to apply Dispersal-Extinction-Cladogenesis (DEC) analyses to obtain biogeographic history inferences. The result showed species of the genera Hyperlaelaps and Haemolaelaps and some species of the genus Androlaelaps in the family Laelapidae were divided into clades of the genus Laelaps in both the BI and ML trees. Divergence time estimates and biogeographic history analysis revealed that the family Laelapidae likely diverged from other taxa during the Middle Jurassic (ca. 156.73 Mya), with Asia considered the most likely ancestral region for the family Laelapidae. Species of various genera began to undergo massive diversification events during the Cenozoic Tertiary. The results suggest that some genera in the family Laelapidae need to be re-defined or new genera need to be established; the Late Cretaceous to Late Neogene warm period would have promoted the divergence and expansion of species in the family Laelapidae. The divergence and dispersal of the family Laelapidae species is most likely a joint response to the continued northward drift of the Indian plate away from the Gondwana paleo-continent and gradually closer to Asia during the Late Cretaceous and the geological activity of the Tibetan Plateau during the Cenozoic Tertiary. The results strengthen our understanding of the origin and evolution of species in the family Laelapidae.

Inference methods for the Very Flexible Weibull distribution based on progressive type-II censoring

In this work, we present classical and Bayesian inferential methods based on samples in the presence of progressive type-II censoring under the Very Flexible Weibull (VFW) distribution. The considered distribution is relevant because it is an alternative to traditional non-flexible distributions and also to some flexible distributions already known in the literature, keeping the low amount of two parameters. In addition, studying it in a context of progressive censoring allows attesting to its applicability in data modeling from various areas of industry and technology that can use this censoring methodology. We obtain the maximum likelihood estimators of the model parameters, as well as their asymptotic variation measures. We propose the use of Markov chain Monte Carlo methods for the computation of Bayes estimates. A simulation study is carried out to evaluate the performance of the proposed estimators under different sample sizes and progressive type-II censoring schemes. Finally, the methodology is illustrated through three real data sets.

Data fusion method for wireless sensor network based on machine learning

In order to improve the energy consumption balance between wireless sensor nodes and reduce the energy consumption of nodes in the process of data fusion, a machine learning based data fusion method for wireless sensor networks is proposed. Through the establishment and training of wireless sensor network model, the compressed sensing method is used to collect wireless sensor network data, and the multi-dimensional de aggregation class analysis algorithm is used to de duplicate the collected data. Using the spatial correlation between the data collected by multiple sensor nodes, the DCS method is used to process the abnormal data of WSN network. In order to eliminate the influence of measurement error on the fusion accuracy, the WSN network data is preliminarily fused by combining the adaptive theory with the batch estimation fusion algorithm. Based on the preliminary fusion results of WSN network data, the Bayesian inference method in machine learning algorithm is used to further fuse WSN network data. The experimental results show that the number of surviving nodes is large and the energy consumption is low when using this method for data fusion. The energy consumption between wireless sensor nodes has a certain balance, which proves that this method has a good data fusion effect.