Estimation of missing data values during data migration using AQ algorithm in cancer detectionJournal: International Journal for Scientific Research and Development | IJSRD (Vol.3, No. 11)
Publication Date: 2016-02-01
Authors : N.Vikasini; Dr.J.Shanthini;
Page : 841-844
Keywords : Data Mining; Decision Tree; Missing value; Modified AQ Algorithm; Classification;
In Data Mining, data missing is more complicated problem now days, especially in hospitals. Most of the hospitals are using client server technology for data transferring inside the hospital. When migrating database from access to SQL, or Migrating data from SQL to Oracle, data loss may occur. Due to the data loss, some values from the table or from database may disappear. This problem is known as Data Missing. Cancer which is the deadliest disease in which if data is missing including their infection percentage along with the missing values in the database leads to a serious problem. To find out the missing values, sometimes prediction may be used to fill the data. Prediction should me more accurate. So, here we are implementing a multidimensional array model with modified AQ algorithm. An improved AQ algorithm is used to find the missing values with 9.5.79% accuracy in the dataset. From the data set, an operational database will be created for the cancer patients and a database for normal patients. This database will be unique and different types of sample data are available. The modified AQ algorithm will compare the existing spatial database with the normal database from the input database. So that the result will be obtained from the dataset, whether the patient is affected from cancer or not, including their infection percentage along with the missing values in the database.
Other Latest Articles
Last modified: 2016-02-12 18:39:29