Document Details

Document Type : Thesis 
Document Title :
A Comparison of Parametric and Non Parametric Methods for the Effectiveness of Wet Cupping Therapy
مقارنة بين الطرق المعلمية واللامعلمية لتأثير العلاج بالحجامة الرطبة
 
Subject : Faculty of Science 
Document Language : Arabic 
Abstract : Data mining is the process of noticing unknown and unseen useful information, it is aimed to gain beneficial and understandable information from the huge data sets. The classification is a supervised method used to classify the data by learning a part of it as sample then predicts other data to test the performance. The interpretation of data and the analysis are the most important parts in medical researches. This thesis evaluates the possibility of using both parametric and nonparametric methods in classification and prediction. This retrospective study applied on patients who did wet cupping, and objects to investigate the effect of wet cupping therapy (WCT) on some diagnosis using an artificial neural network, decision tree, discriminant analysis and logistic regression model. The proposed models determined the important variables for each data set and employed as a systems that give an accurate, inexpensive and suitable explanation to predict the effect of WCT on the patient additionally to forecast level of improvement. These models will help the practitioner to take the right decision of predicting the patients’ response for WCT. The investigated data set that have major impact on predicting the effect of WCT are liver function, the problems of cells that circulate in blood, thyroid function, kidney function, lipid, bone and calcium problems. The models constructed by dividing the data into 70% to train the model and 30% to test their performance using different classification assessments. The results of these models present the importance of WCT on patient’s health depending on the diagnosis, where there are significant difference on most of blood tests. The comparisons between models were made to identify a comprehensive evaluation about the models. The study conclude that WCT is an effective treatment in improving the quality of life and reducing pain. The proposed models identify the important features that have a major influence in each data set. These models will help the clinicians to predict the level of patients’ improvement. Any model which proposed could be applied as a supportive tool to make an early decision regarding the response of patient to the WCT. It would be helpful to apply a nonparametric method with a parametric method and identify the performance of each methods to select the optimal one for classification. The results would provide guidelines for the improvement levels by targeting some types of blood tests and select better method in prediction power. 
Supervisor : Dr. Lamya Abdulbaset Baharith 
Thesis Type : Master Thesis 
Publishing Year : 1442 AH
2020 AD 
Added Date : Sunday, January 17, 2021 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
هاجر صالح الجهنيAljohani, Hajar SalehResearcherMaster 

Files

File NameTypeDescription
 46850.pdf pdf 

Back To Researches Page