Obsah: :: Library Catalog

Uloženo v:

Podrobná bibliografie
Hlavní autoři:	Prajna Bhunia, Sirsendu Das Adhikary, Supriya Maity, Dipankar Dey, Samiram Pal
Médium:	Recurso digital
Jazyk:	angličtina
Vydáno:	Zenodo 2023
Témata:	Machine Learning Linear Regression Support Vector Machine Neural Network Gaussian Process Regression Tree Urine pH
On-line přístup:	https://doi.org/10.5281/zenodo.14740345
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Obsah:

<div> <h1>A comparative analysis of multiple approaches machine learning for predicting and analysing urine pH amount</h1> <h2>aPrajna Bhunia , bSirsendu Das Adhikary, cSupriya Maity, dDipankar Dey, eSamiram Pal</h2> abcdeGlobal Institute of Science & Technology, Haldia, Purba Midnapur-721657, West Bengal, India Email: <a href="mailto:sirsendu1979@gmail.com">sirsendu1979@gmail.com,</a> <a href="mailto:supriyamaity1234@gmail.com">supriyamaity1234@gmail.com,</a> deydipankar2014@gmail.com, samiran.sip@gmail.com   </div> ABSTRACT Prenatal treatment includes clinical urine testing as a crucial element.Medical professionals now evaluate urine test strips using an operator-dependent, labor-intensive, and visually color-coded process that takes a long time. Procedures and methods: By using various treatment and resource recovery techniques, urine has the potential to offer numerous useful resources. Selecting which technique to utilize and what re- sources might be retrieved from human urine, we paid particular attention to pH because it was thought to be the most significant parameter. We made a distinction between fresh, hydrolyzed, and stabilized urine treat- ment methods. For optimum resource recovery, future studies should concentrate on a thorough economic and life-cycle assessment of the urine treatment process. It has been shown that ML and AI are beneficial in a variety of fields, particularly with the current explosion of data. Making quicker and more accurate judgments in terms of illness forecasts may be possible using this method. Machine learning algorithms are therefore increasingly being used in prediction applications. Because of its high degree of accuracy, ML has been adopted by clinical diagnostics as one of the main computational approaches and analytics for illness identification. In order to increase the consistency and quality of disease reporting, building a model can also help us visualize and analyze diseases. This article has investigated how to predict the average pH value of urine. Different ML algorithms, including Linear Regression, Support Vector Machine, Neural Network, Gaussian Process Regression, and Fine Tree, are used to learn and find meaningful patterns. There are sev- eral insightful discoveries in this article. The R2 number is used to assess the accuracy of machine learning methods, including Fine Tree, Gaussian Process Regression, Neural Network, Support Vector Machine, and Linear Regression. According to recent research, the Linear Regression algorithm has the lowest RMSE value when compared to other algorithms and a high accuracy rate of 0.99997 for R2. Nevertheless, the difficult and future research area for these studies will be to raise the accuracy rates of the machine learning algorithms.

Podobné jednotky