![](/images/graphics-bg.png)
An enhanced algorithm with modified multiple minimum support using classification algorithm
Other Title(s)
تحسين و تعديل خوارزمية متعددة قيم الحد الأدنى للدعم باستخدام خوارزمية التصنيف
Dissertant
Thesis advisor
Comitee Members
Abu Ata, Bilal M.
Alawinah, Ali. A.
University
Philadelphia University
Faculty
Faculty of Information Technology
Department
Department of Computer Science
University Country
Jordan
Degree
Master
Degree Date
2016
English Abstract
Data mining is a process of extracting desirable knowledge or interesting patterns from existing databases for specific purposes.
Previous approaches used both a single minimum support that indicated for all items or item sets, and multiple minimum support which gave one value for each item.
But in real applications, different items may have different criteria to judge its importance, so multiple minimum support is more useful than single minimum support because it is more justifiable method to use and it gave us more real results, that’s why all new associations rule used this new method.
Multiple minimum support faces some of challenges, one of these challenges it can't deal with the rare items, the other challenge it can't deal with too high or too low multiple minimum support values.
In this thesis we will solve the weaknesses of using multiple minimum support, we will provide another point of view about defining the multiple minimum support values by modifying a new classification algorithm methodology, this modified algorithm will divide data to classes according to a unique relationships between items, then compute the average of all items in each class and give the new value for each item in one class, this will be applied for all items in one class.
As a result, to a void dropping in rare items problem and get high quality results.
The proposed algorithm it is easy and efficient, it deals with rare items, huge data, too low and too high multiple minimum support values.
Furthermore, it will give us high quality and more flexible results.
Classification is a process used for portioning the data into different classes according to some of constrains.
In other words we can say classification is a process of generalizing the data according to different instances, we will use classification to assign items to classes.
The modified algorithm will be implemented by improving new design methodology to multiple minimum support in new way.
This modified algorithm will give extra efficiency to using multiple minimum support.
Main Subjects
Information Technology and Computer Science
No. of Pages
40
Table of Contents
Table of contents.
Abstract.
Abstract in Arabic.
Chapter One : Introduction.
Chapter Two : Background of multiple minimum support.
Chapter Three : Approaches of multiple minimum support.
Chapter Four : Implementation details classes and features.
Chapter Five : Implementation issues, evaluation and application areas.
References.
American Psychological Association (APA)
al-Rawashidah, Muhammad Akram. (2016). An enhanced algorithm with modified multiple minimum support using classification algorithm. (Master's theses Theses and Dissertations Master). Philadelphia University, Jordan
https://search.emarefa.net/detail/BIM-725357
Modern Language Association (MLA)
al-Rawashidah, Muhammad Akram. An enhanced algorithm with modified multiple minimum support using classification algorithm. (Master's theses Theses and Dissertations Master). Philadelphia University. (2016).
https://search.emarefa.net/detail/BIM-725357
American Medical Association (AMA)
al-Rawashidah, Muhammad Akram. (2016). An enhanced algorithm with modified multiple minimum support using classification algorithm. (Master's theses Theses and Dissertations Master). Philadelphia University, Jordan
https://search.emarefa.net/detail/BIM-725357
Language
English
Data Type
Arab Theses
Record ID
BIM-725357