Improve the classification and Sales management of products using multi-relational data mining
There are some elements such as competition among companies and changes in demands which result in changes of customers behaviors. Therefore, paying no attention to these changes may lead to a reduction in company benefits and loss of customers. Since data and their analyses determine the activities and decision makings of companies, data quality is of paramount in analyzing them because misinformation leads to wrong decision making. Since data mining has been designed to find out multi repetition patterns, it can be used to improve the product sales violations by sales people and increase the quality of data. Most of data mining models available try to find patterns in one table, but the fact is that standard database is of more applications because they are normalized and the data are scattered in more than one table. This situation calls for a new method to discover the data and interpret them. This paper tries to discuss this issue as the MultiRelational Data Mining (MRDM). In fact, this MRDM is used to improve product sales and categories. This method, indeed, enables us to extract the knowledge beyond several tables in standard database and compare data with the extracted patterns. There is a method to measure and evaluate data from some tables in the database with the approach of MRDM.
Multi Relational Data Mining Data Quality
Mahmoud Houshmand Mohammad Alishahi
Industrial Engineering Department Sharif University of Technology, Azadi Ave.Tehran, IRAN Industrial Engineering Department, Sharif University of Technology, Azadi Ave.Tehran, IRAN
国际会议
贵阳
英文
329-337
2011-01-26（万方平台首次上网日期，不代表论文的发表时间）