Bayesian and Decision Tree Approaches for Pattern Recognition Including Feature Measurement Costs

Dattatreya, G.R.; Sarma, V. V. S.

doi:10.1109/tpami.1981.4767102

Cited by 30 publications

(10 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, in entry level of the industry, e-health sensors will gather all sensors data which will used for training our system. In our proposed system, a probabilistic model called Bayesian Decision Tree [57,58] has been used for training and validation to get the health condition score for fitness of the employee. We have used this mixture model of decision tree and Bayesian theory to handle categorical variables and missing data in principled way to get better performance.…”

Section: Health Condition Detectionmentioning

confidence: 99%

<em>i</em>WorkSafe: Towards Healthy Workplaces during COVID-19 with an Intelligent pHealth App for Industrial Settings

Kaiser¹,

Mahmud²,

Noor³

et al. 2021

Preprint

View full text Add to dashboard Cite

The recent outbreak of the novel Coronavirus Disease (COVID-19) has given rise to diverse health issues due to its high transmission rate and limited treatment options. Almost the whole world, at some point of time, was placed in lock-down in an attempt to stop the spread of the virus, with resulting psychological and economic sequela. As countries start to ease lock-down measures and reopen industries, ensuring a healthy workplace for employees has become imperative. Thus, this paper presents a mobile app-based intelligent portable healthcare (pHealth) tool, called iWorkSafe, to assist industries in detecting possible suspects for COVID-19 infection among their employees who may need primary care. Developed mainly for low-end Android devices, the iWorkSafe app hosts a fuzzy neural network model that integrates data of employees' health status from the industry's database, proximity and contact tracing data from the mobile devices, and user-reported COVID-19 self-test data. Using the built-in Bluetooth low energy sensing technology and K Nearest Neighbor and K-means techniques, the app is capable of tracking users’ proximity and trace contact with other employees. Additionally, it uses a logistic regression model to calculate the COVID-19 self-test score and a Bayesian Decision Tree model for checking real-time health condition from intelligent e-health platform for further clinical attention of the employees. Rolled out in an apparel factory on 12 employees as a test case, the pHealth tool generates an alert to maintain social distancing among employees inside the industry. In addition, the app helps employee to estimate risk with possible COVID-19 infection based on the collected data and found that the score is effective in estimating personal health condition of the app user.

show abstract

Section: Health Condition Detectionmentioning

confidence: 99%

<em>i</em>WorkSafe: Towards Healthy Workplaces during COVID-19 with an Intelligent pHealth App for Industrial Settings

Kaiser¹,

Mahmud²,

Noor³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Each node in the output layer is associated with each sample label. ID3 can be acquired according to Dattatreya (Dattatreya and Sarma 1981), but this method is redundant to calculated posterior probability. Thus, steps of simplifying the tree and enhancing its generalization ability are performed using post-pruning algorithm (Murthy et al 2001).…”

Section: Posterior Probability Computationmentioning

confidence: 99%

“…To determine the optimal K 1 and K 2 , the misclassification rates on the whole training samples are estimated. Step2 Compute K 1 × K 2 misclassification rates from the combination of the K 1 classifiers with K 2 features on (Paliwal and Rao 1983) pp ¼ l p k ; l p > l n À Á l n k ; l p ≤l n À Á 2 6 4 l p : numbers of positive feature points l n : numbers of negative feature points k: total numbers of feature points LDA (Martinez and Kak 2001) (Dattatreya and Sarma 1981) …”

Section: Improved Majority Rulementioning

confidence: 99%

Identification of Impurities in Fresh Shrimp Using Improved Majority Scheme-Based Classifier

2016

View full text Add to dashboard Cite

The efficient removal of impurities from postharvest raw shrimp is beneficial to improve the quality of shrimp products. Single feature combined with single classifier results in poor classification rate. Moreover, accuracy of combined classifiers remains unsatisfactory, especially when sound shrimp is mixed with various defective shrimp and impurities. In this study, an improved majority rule (IMAJ) classifier combination scheme was proposed to address this problem. The accuracy of IMAJ (91.53 %) was compared with six other kinds of classifier combination schemes proposed by Kittler. The schemes include Sum (89.93 %), Product (65.99 %), Max (89.76 %), Min (80 %), Median (88.18 %), and Majority (89.2 %). Comparison results indicate that the combination classifier based on IMAJ rule is superior.

show abstract

“…This will be available with the system. Suppose that the performance constraints available from the analysis of the problem give the following matrix for cost of misclassification : [12] can be used to get the required decision tree. Figure 2 gives the resultant decision tree [12].…”

Section: System For Design Of Simple Pr Problemsmentioning

confidence: 99%

A Knowledge-Based Design Environment for Speech Recognition Systems

Chitnis

Sarma

1988

IETE Journal of Research

View full text Add to dashboard Cite

The progress in speech technology over the last 30 years has come to a stage, where simple yet practical speech recognition systems could be built. Designing of such systems is an evolutionary process requiring a lot of experimentation. This lengthy and involved process is a major hurdle in developing useful commercial systems like a ,·oice-input telephone dialer. To overcome this problem, we propose the use of a knowledge-based design environment.For analysing a given problem, a framework based on PR and AI concepts is explained using the example of a telephone dialer. Then some representative speech recognition/understanding systems are reviewed to give some idea about the design aspects like the system architecture, PR subtasks involved, AI techniques used etc. Finally the concept of a design environment is explained. The environment is considered as a collection of knowledge-based tools with a global knowledge-base about speech technology.

show abstract

Bayesian and Decision Tree Approaches for Pattern Recognition Including Feature Measurement Costs

Cited by 30 publications

References 10 publications

<em>i</em>WorkSafe: Towards Healthy Workplaces during COVID-19 with an Intelligent pHealth App for Industrial Settings

<em>i</em>WorkSafe: Towards Healthy Workplaces during COVID-19 with an Intelligent pHealth App for Industrial Settings

Identification of Impurities in Fresh Shrimp Using Improved Majority Scheme-Based Classifier

A Knowledge-Based Design Environment for Speech Recognition Systems

Contact Info

Product

Resources

About