Understanding the effects of variation of parameters (like advertisements, impressions, cost, click conversion rate, etc.) as the deep learning model for predicting sales revenue starts to iterate through the dataset. The ODE of various orders governs this variation of parameters. We have built and trained an Artificial Neural Network (ANN) that uses Hessian matrix optimisation techniques to find the most optimum way of increasing a company’s sales revenue.
There were many simple things that we would have done. But, we thought let’s push ourselves a step further and go, research, and learn on a completely new topic. We are glad that we have learnt something different from the textbooks and thoroughly understood and enjoyed the concept which is quite widespread nowadays. We would like to thank the professors for giving us an opportunity to do such a project.
The code is licenced under the MIT license and free to use by anyone without any restrictions.
Created with :heart: by Mumuksh Tayal