Estimating the strength of concrete can help in developing safety guidelines governing the use in the construction of buildings. In the current project I have tried to create a model that can accurately predict the concrete strength given the composition of input materials. To do so, I have tried to replicate I-Cheng Yeh’s data and results using a simple neural network model in R.
URL : http://archive.ics.uci.edu/ml
Further Information : Modeling of strength of high performance concrete using artifiial neural networks. Cement and Concrete Research. 1998; 28:1797-1808.
By deploying a very simple Neural Network model I was able to replicate the accuracy reported by Yeh in his paper. This demonstrates the applicability and ease with which neural networks can showcase high performance. To optimize, I had used varied degree of threshold and hidden layers and achieved a correlation of 0.82745.