Generalized posterior probability for minimizing verification errors at subword, word and sentence levels

Author(s):  
Wai Kit Lo ◽  
F.K. Soong ◽  
S. Nakamura
1990 ◽  
Vol 29 (03) ◽  
pp. 167-181 ◽  
Author(s):  
G. Hripcsak

AbstractA connectionist model for decision support was constructed out of several back-propagation modules. Manifestations serve as input to the model; they may be real-valued, and the confidence in their measurement may be specified. The model produces as its output the posterior probability of disease. The model was trained on 1,000 cases taken from a simulated underlying population with three conditionally independent manifestations. The first manifestation had a linear relationship between value and posterior probability of disease, the second had a stepped relationship, and the third was normally distributed. An independent test set of 30,000 cases showed that the model was better able to estimate the posterior probability of disease (the standard deviation of residuals was 0.046, with a 95% confidence interval of 0.046-0.047) than a model constructed using logistic regression (with a standard deviation of residuals of 0.062, with a 95% confidence interval of 0.062-0.063). The model fitted the normal and stepped manifestations better than the linear one. It accommodated intermediate levels of confidence well.


1990 ◽  
Vol 17 (27) ◽  
pp. 59-73 ◽  
Author(s):  
Shin’ichi Ichikawa ◽  
Hiroshige Takeichi

Entropy ◽  
2021 ◽  
Vol 23 (8) ◽  
pp. 983
Author(s):  
Jingjian Li ◽  
Wei Wang ◽  
Hong Mo ◽  
Mengting Zhao ◽  
Jianhua Chen

A distributed arithmetic coding algorithm based on source symbol purging and using the context model is proposed to solve the asymmetric Slepian–Wolf problem. The proposed scheme is to make better use of both the correlation between adjacent symbols in the source sequence and the correlation between the corresponding symbols of the source and the side information sequences to improve the coding performance of the source. Since the encoder purges a part of symbols from the source sequence, a shorter codeword length can be obtained. Those purged symbols are still used as the context of the subsequent symbols to be encoded. An improved calculation method for the posterior probability is also proposed based on the purging feature, such that the decoder can utilize the correlation within the source sequence to improve the decoding performance. In addition, this scheme achieves better error performance at the decoder by adding a forbidden symbol in the encoding process. The simulation results show that the encoding complexity and the minimum code rate required for lossless decoding are lower than that of the traditional distributed arithmetic coding. When the internal correlation strength of the source is strong, compared with other DSC schemes, the proposed scheme exhibits a better decoding performance under the same code rate.


2013 ◽  
Vol 838-841 ◽  
pp. 1463-1468
Author(s):  
Xiang Ke Liu ◽  
Zhi Shen Wang ◽  
Hai Liang Wang ◽  
Jun Tao Wang

The paper introduced the Bayesian networks briefly and discussed the algorithm of transforming fault tree into Bayesian networks at first, then regarded the structures impaired caused by tunnel blasting construction as a example, introduced the built and calculated method of the Bayesian networks by matlab. Then assumed the probabilities of essential events, calculated the probability of top event and the posterior probability of each essential events by the Bayesian networks. After that the paper contrast the characteristics of fault tree analysis and the Bayesian networks, Identified that the Bayesian networks is better than fault tree analysis in safety evaluation in some case, and provided a valid way to assess risk in metro construction.


2014 ◽  
Vol 100 (1) ◽  
pp. 14-17 ◽  
Author(s):  
Syed Mohinuddin ◽  
Pankaj Sakhuja ◽  
Benjie Bermundo ◽  
Nandiran Ratnavel ◽  
Stephen Kempley ◽  
...  

Bilious vomiting in a neonate may be a sign of intestinal obstruction often resulting in transfer requests to surgical centres. The aim of this study was to assess the use of clinical findings at referral in predicting outcomes and to determine how often such patients have a time-critical surgical condition (eg, volvulus, where a delay in treatment is likely to compromise gut viability).Methods4-year data and outcomes of all term newborns aged ≤7 days with bilious vomiting transferred by a regional transfer service were analysed. Specificity, sensitivity, likelihood ratios, correlations, prior and posterior probability of clinical findings in predicting newborns with surgical diagnosis were calculated.ResultsOf 163 neonates with bilious vomiting, 75 (46%) had a surgical diagnosis and 23 (14.1%) had a time-critical surgical condition. The diagnosis of a surgical condition in neonates with bilious vomiting was significantly associated with abdominal distension (χ2=5.17, p=0.023), abdominal tenderness (χ2=5.90, p=0.015) and abnormal abdominal X-ray findings (χ2=5.68, p=0.017) but not with palpation findings of a soft as compared with a tense abdomen (χ2=3.21, p=0.073). Abnormal abdominal X-ray, abdominal distension and tenderness had 97%, 74% and 62% sensitivity, respectively, with regard to association with an underlying surgical diagnosis. Normal abdominal X-ray reduced the posterior probability of surgical diagnosis from 50% to 16%. Overall, clinical findings at referral did not differentiate between infants with or without surgical or time-critical condition.ConclusionsWe recommend that term neonates with bilious vomiting referred for transfer are prioritised as time critical.


Geophysics ◽  
2013 ◽  
Vol 78 (3) ◽  
pp. V87-V100 ◽  
Author(s):  
Caglar Yardim ◽  
Peter Gerstoft ◽  
Zoi-Heleni Michalopoulou

Sequential Bayesian techniques enable tracking of evolving geophysical parameters via sequential observations. They provide a formulation in which the geophysical parameters that characterize dynamic, nonstationary processes are continuously estimated as new data become available. This is done by using prediction from previous estimates of geophysical parameters, updates stemming from physical and statistical models that relate seismic measurements to the unknown geophysical parameters. In addition, these techniques provide the evolving uncertainty in the estimates in the form of posterior probability density functions. In addition to the particle filters (PFs), extended, unscented, and ensemble Kalman filters (EnKFs) were evaluated. The filters were compared via reflector and nonvolcanic tremor tracking examples. Because there are numerous geophysical problems in which the environmental model itself is not known or evolves with time, the concept of model selection and its filtering implementation were introduced. A multiple model PF was then used to track an unknown number of reflectors from seismic interferometry data. We found that when the equations that define the geophysical problem are strongly nonlinear, a PF was needed. The PF outperformed all Kalman filter variants, especially in low signal-to-noise ratio tremor cases. However, PFs are computationally expensive. The EnKF is most appropriate when the number of parameters is large. Because each technique is ideal under different conditions, they complement each other and provide a useful set of techniques for solving sequential geophysical inversion problems.


Sign in / Sign up

Export Citation Format

Share Document