期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2015
卷号：6
期号：3
页码：2574-2578
出版社：TechScience Publications
摘要：Data mining technique has seen the tremendous increase in their application recently. The biggest challenge has been related to the creation of meaningful data model and an iterative process of using data transformation mechanism. These challenges are coupled with the unsolved problems that arise while treating data that results in high inaccuracy by using conventional methods. Huge amount of data is generated in health care transaction which is increasingly becoming difficult to understand, manage and analyse manually. In order to analyse clinical data effectively and keep up with the technology producing high intensity data, Machine Learning techniques are involved to solve this complex problem. A machine learning algorithm learns from the previous experience and automatically takes decisions. Gene Expression levels change over time. These profiles can distinguish between cells that are actively dividing, or show how the cells react to a particular treatment. The knowledge acquired though the data analysis can’t be handled explicitly for future use. They may also contain missing data value which makes analysis more difficult. Therefore sets of genes that are co-ordinately expressed in different diseases and defined as specific modules, often demonstrating coherent functional relationships through unbiased literature profiling. There are many papers who have applied different data mining and machine learning techniques to solve these challenges. This paper highlights the main challenges involved in analysing and identifying the patients and classifying them into HIV with TB and those having HIV without traces of TB disease. This paper is a brief survey on the techniques used to perform the classification of HIV/TB co-infected patients.
