Skip to Main Content
A common trend in several areas of knowledge is to combine distinct methods or algorithms expertise in order to provide more accurate results. This approach has its basis on the theory of wisdom of crowds, which claims that the information drawn from collective decisions is usually more precise than the individual ones. Many classification tasks have profit from this idea. However, its application to gene regulatory networks inference is recent and still not deeply explored. In the current work, we perform a comparative study between several widely used combination methods in machine learning. We analyze their performance for artificially generated gene networks and observe that ensemble predictions yield more accurate results than individual ones, thus being an interesting strategy for improving inference on gene regulatory networks.