On the dependability of experimental data in bio-medical explorations

The Big Data era has introduced many complications concerning the quality of information and experiments due to several factors considered in this research. Microarray technology is one of the tools that generate extensive amounts of biological data. The results of this technology can help in finding patterns and relations within biological data, in addition to assisting with the development of diagnostic systems. This research uses Weka to apply multiple machine-learning algorithms to a Leukemia dataset that was produced using microarray technology. The results of this data mining application and its underlying concerns are also discussed. Furthermore, it is clear that irreproducible biomedical studies are at a high level of magnitude, which is problematic as these studies serve as the infrastructure of information for other research. The placebo effect is also discussed in this paper as it poses a threat to the quality of biological information. Additionally, biomedical outcomes could contribute to skepticism about the effectiveness of "Evidence-based Medicine," which will be discussed in this research. Moreover, this research discusses some unexpected results of biological experiments; these results are then directly linked to the complex and extensive amount of biomedical data. In addition, I claim that the results of applying data mining methods to this data would generate erroneous and inaccurate conclusions and decisions. One of the promising applications for data mining towards biomedical computing, which exploits users' search queries to test the effect of adverse drugs based on real time, is also discussed briefly.