Understanding the data and reaching accurate conclusions are of paramount importance in the present era of big data. Machine learning and probability theory methods have been widely used for this purpose in various fields. One critically important yet less explored aspect is capturing and analyzing uncertainties in the data and model. Proper quantification of uncertainty helps to provide valuable information to obtain accurate diagnosis. This paper reviewed related studies conducted in the last 30 years (from 1991 to 2020) in handling uncertainties in medical data using probability theory and machine learning techniques. Medical data is more prone to uncertainty due to the presence of noise in the data. So, it is very important to have clean medical data without any noise to get accurate diagnosis. The sources of noise in the medical data need to be known to address this issue. Based on the medical data obtained by the physician, diagnosis of disease, and treatment plan are prescribed. Hence, the uncertainty is growing in healthcare and there is limited knowledge to address these problems. Our findings indicate that there are few challenges to be addressed in handling the uncertainty in medical raw data and new models. In this work, we have summarized various methods employed to overcome this problem. Nowadays, various novel deep learning techniques have been proposed to deal with such uncertainties and improve the performance in decision making.

/pdf/handling-of-uncertainty-in-medical-data-using-machine-rpi8bge796.pdf

Handling of uncertainty in medical data using machine learning and probability theory techniques: a review of 30 years (1991–2020)

Understanding data and reaching valid conclusions are of paramount importance in the present era of big data. Machine learning and probability theory methods have widespread application for this purpose in different fields. One critically important yet less explored aspect is how data and model uncertainties are captured and analyzed. Proper quantification of uncertainty provides valuable information for optimal decision making. This paper reviewed related studies conducted in the last 30 years (from 1991 to 2020) in handling uncertainties in medical data using probability theory and machine learning techniques. Medical data is more prone to uncertainty due to the presence of noise in the data. So, it is very important to have clean medical data without any noise to get accurate diagnosis. The sources of noise in the medical data need to be known to address this issue. Based on the medical data obtained by the physician, diagnosis of disease, and treatment plan are prescribed. Hence, the uncertainty is growing in healthcare and there is limited knowledge to address these problems. We have little knowledge about the optimal treatment methods as there are many sources of uncertainty in medical science. Our findings indicate that there are few challenges to be addressed in handling the uncertainty in medical raw data and new models. In this work, we have summarized various methods employed to overcome this problem. Nowadays, application of novel deep learning techniques to deal such uncertainties have significantly increased.

Handling of uncertainty in medical data using machine learning and probability theory techniques: A review of 30 years (1991-2020)

Data Mining is very useful and widely applied in various fields, one of which is in the field of education. Data mining can be applied to the field of Educational institutions and is also often referred to as Educational Data Mining (EDM). Colleges currently compete with each other to improve the quality of their education for the creation of competent and high-integrity human resources. The amount of data contained in Higher Education can be utilized for the need for useful information so that the data attributes can be known so that the data is analyzed to improve student performance and achievement. And also the results of the analysis are expected to be able to anticipate the problem of delays in the study period that is often experienced by students. In this study conducted using two algorithm models namely K-Nearest Neighbor and Decision Tree C4.5. The best accuracy value is the K-Nearest Neighbor algorithm model with an accuracy rate of 59.32%, whereas in the Decision Tree C4.5 model the accuracy rate is 54.80%, the application of EDM and is expected to be maximized and developed so that it can contribute and develop in education world especially in data mining

/pdf/implementation-educational-data-mining-for-analysis-of-4800j7nd76.pdf

Implementation Educational Data Mining For Analysis of Student Performance Prediction with Comparison of K-Nearest Neighbor Data Mining Method and Decision Tree C4.5: Implementation Educational Data Mining For Analysis of Student Performance Prediction with Comparison of K-Nearest Neighbor Data Mining Method and Decision Tree C4.5

Many person are not aware of the dangers of depression. It's hard to know the symptoms and overcome depression early on. In this case adolescents are the most vulnerable to the disease because at that time humans are in a transition phase from childhood to adulthood. WHO says every second there is a case of suicide due to severe depression. Based on these cases the researchers aimed to make an application to help the community, especially students, to be able to recognize and diagnose depression from an early age. The system in this application is an expert system using the Certainty Factor and Forward Chaining method. Inputs received by this application are biodata and psychological questions about the symptoms experienced by the user. The resulting output is the diagnosis, as well as the percentage of possible diseases suffered by user.

Diagnosa depresi pada mahasiswa menggunakan metode certainty factor dan forward chaining

Currently, the Micro, Small and Medium Enterprises (MSMEs) in Banten Indonesia mostly still manage their customers in conventional ways, the touch of information technology has not been done much. One approach to modern business is to pay attention to the factors of good relationships with customers, not only before the transaction, but also after the transaction. The Information Technology System used to manage customer relationships with the aim of increasing customer retention is called Customer Relationship Management (CRM). The existing CRM software is more oriented to large companies that are already established, so that it will cause some discrepancies when applied in MSMEs. This study tries to develop E-CRM software that is tailored to the business needs of MSMEs in Banten by using a local approach to facilitate users in their use. This feature in the ECRM will help MSMEs in implementing customer relationship management so that they can increase their competitiveness with other companies in maintaining the business they are running and as a means to obtain a wider new market. The approach used by local policy is through observations made on MSMEs in the Banten region. The application developed in this paper is based on 9 characteristics of the Banten MSMEs produced in previous studies [1] and the E-CRM UMKM blue print model that has been published in international seminars [2]. This application development used the stages that exist in the waterfall model method, namely planning, analysis, design, and implementation. The results of the E-CRM UMKM Application are expected to increase customer retention and build global marketing for MSMEs in the Banten region of Indonesia.

/pdf/electronic-customer-relationship-management-e-crm-3lbltodxci.pdf

Electronic Customer Relationship Management (E-CRM) Application as Efforts to Increase Customer Retention of Micro Small and Medium Enterprises (MSMEs) in Banten Indonesia

Digestive system disorders are dangerous diseases. Lack of awareness of public health is still low, life habits, behavior and mindset that want to live practically, a means of delivering information about diseases that are still lacking, and the lack of medical personnel is a problem, therefore an expert system application is needed to diagnose disease on the web-based digestive system. By providing certainty in the form of a percentage, then using the calculation through symptoms chosen by the user of each symptom has a density value, the density value is obtained from the results of interviews with doctors. Web-based E-Diagnostics for digestive system disorders uses the Dempster Shafer method which is expected to help users by providing information on disease diagnosis and solutions that can be done to help cure it, the Dempster Shafer Method has the ability to provide a high level of accuracy or certainty, which method this has characteristics that are in accordance with the way an expert thinks. This web-based expert system application will display symptoms that can be selected by the user to get the final results in the form of rapid disease diagnoses and suggestions for prevention in order to find out information in the form of diagnoses of diseases of the digestive system. Based on the calculation of accuracy that has been done in this study, it can be seen that the Dempster Shafer Method is the method that has the highest value with 85% confidence compared to the Certainty Factor Method with a value of 60%.

/pdf/web-based-e-diagnostic-for-digestive-system-disorders-in-2soe08l064.pdf

Eka Ayunda Prameswari

Papers

Web-based E-diagnostic for Digestive System Disorders in Humans using the Demster Shafer Method