Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 42674 (01)
- University: Sharif University of Technology
- Department: Industrial Engineering
- Advisor(s): Hooshmand, Mahmoud
- Abstract:
- Nowadays, because of high volume and growth of data in industrial organizations and productive factories, registration and storing of data have forgotten manual and tradition styles for which using automation and mechanized machinery and systems has been a necessary task. In order to reach to this revolution, need to some tools, facilities and methods which can fulfill this requirement is felt strongly. Therefore, high volume of data is considered as an advantage because based on precise analysis it is possible to make logical management decisions with less risk. During last years, statistical and numerical methods and simulation were used to discover knowledge and information when one of the most interesting methods for researchers and managers was using data mining methods. In this method, patterns and hidden knowledge in bases are discovered by using data mining algorithms. In the most of the methods presented for data mining, most of the focus is emphasized on a table of data base, then related information and patterns of a table are extracted. These methods cannot be always effective and useful because in standard databases, there are different tables which are related to each other and data is distributed in several tables. Therefore, to analyze data, multi relation Data Mining is required to be used in order to analyze data through relations among tables. Extracted information via analysis of high volume data sets can be effective and feasible and help to make decisions if there are low data errors. In fact, decision making should be done based on correct data because tables having wrong data cannot help to extract correct information and if no pay attention to this issue, then industrial companies and productive factories may confront with irrecoverable damages. Therefore, to resolve this problem, multi relational data mining is proposed through which we can extract data error via relational table. By this method, accuracy of data is investigated and, if necessary, error data will be corrected an after data correction, decision making or more precise prediction is possible by data analysis. In this research, by using relational council principles, tacit patterns available in data tables set will be discovered. The data which do not follow from these patterns are marked as data error. Discovered errors are ranked by three proposed methods and their results are evaluated and compared
- Keywords:
- Data Mining ; Data Error ; Multi-Relational Data Mining ; Relational Association Rules
Digital Object List-
محتواي پايان نامه - view
Bookmark