Predictive Modelling & Data Science
Your organization has a need for predictive modelling. You’re trying to detect fraud, predict when customers will cancel their subscription, or figure out when bitcoin will go bust. You need someone who is up to date with the latest and greatest AI techniques, so that they can apply them to these traditional data-science use-cases.
Electric Brain can come in and help you model this data. A typical project looks like this:
First, we just need to find out the business requirements for this model.
What use-case is leading to this requirement?
What is the output from the model going to be used for?
What is the business value of the model?
How accurate does it need to be in order to generate business value? If you have to tradeoff between false positives and false negatives, how much do you want to trade off, and in which direction?
Orientation and access to data
Initially, we just need to get oriented around your data and get direct access to the databases if we can. This may involve getting authorization from your security department. It may also involve sitting down with your database engineers to go over the database structure, with a specific focus on the structure of the tables in question.
If the datasets that you are working with are so large that they need to be processed using special distributed tools, then we will need to get ourselves oriented around your data-processing infrastructure as well.
Now with all of the surrounding stuff done, our team can get down and dirty to the core task of modelling your data.
Time & Cost
The cost of doing this kind of modelling can vary quite a bit, depending on the difficulty of the problem, the size of the dataset, and the desired business objectives and accuracy levels. But to give you an idea, follow these guidelines:
Quick & Dirty
You just need something. The dataset is small, the problem isn’t complicated, and the expectation isn’t high.
Cost: $15,000 to $40,000
Time: 2 weeks to 5 weeks
Maximize the accuracy
You already have a model that is performing well on a difficult problem, but you want to push the limits and get the accuracy even higher. Even a small improvement of 0.5% to 1.0% would be of enormous value to your business.
Cost: $35,000 to $75,000
Time: 10 weeks to 20 weeks