overall, is to anticipate customer behaviors, so offer them possibilities to satisfy their need with our products/services, or giving us the input to create or produce those products/services that this customers segment need.
Hi Jean-Pierre,
For Machine Learning (ML) to make a meaningful contribution, four conditions need to be fulfilled:
- There must be some pattern/correlation between input data, which you have, and an output measurement, which you are trying to predict. You must be able to have some intuition regarding the choice model which “translates” the input into the output.
- You need to have a large body of labeled training data (I limit myself to supervised learning models here)
- You need an algorithm which “trains” the model until it creates the correct outputs from the inputs when used on the training, or better test data.
- Your input data should not be in a range where the relationship between input and output is chaotic for the model chosen.
In areas/challenges where these conditions are fulfilled, there is an almost limitless range of tasks where Machine Learning can be applied. Starting today and speculating about the foreseeable future, I can see the some examples:
- In sales, the qualified leads, as opposed to low-probability opportunities can often be identified
- In online marketing, complex relationships between what is known about a customer and the most successful type of advertising can be deducted
- In legal, peculiar draft contract clauses as well as relevant cases can be identified
- In production, complex construction can be automated
- In procurement, appropriate price-ranges can be determined
- In the supply chain, demand can be projected and distribution, transport and reception can be automated
- In HR, promising candidates can be identified and led through a personalized learning curve for both their wellbeing and the employer’s benefit
- In finance, liquidity can be planned to a higher degree of precision and optimal leverage can be determined
These are just some examples and there are, no doubt, many more. Every one of these probably deserves a book in its own right but I hope that, for now, this will give you some high-level insight. Just comment, if something in particular interests you.
Regards, Eduard
The correct question is “What will not be in the scope of machine learning in the future?”
Haven't found a solution?
This will mark this comment as best reply and close your question.
Are you sure?
This will close your question without a Best reply.
Are you sure?
This will report this content as inappropiate to the moderators.
Are you sure?