You’ve probably heard the words “machine learning” thrown around a lot lately. But what exactly is it and how can it be used to improve your own services or workflows? To demystify this concept, we’ve created a series of articles that look at common problems and the different ways machine learning can be used to solve them. Our first article examines how machine learning can be used to improve customer service.
Let’s say you run a company that gets a wide variety of emails to your customer service account. There are numerous ways you can go about using these emails to make inferences about how your customers are feeling about your business. Many companies have a customer service representative manually read and categorize each and every email. But as the volume of emails you receive increases, this approach can be difficult and time consuming. The following is a look at the various ways companies are tackling this problem, and how machine learning presents a solution.
.post-content #google-cloud-embed iframe {
max-height: none;
}
Although we presented a simplified view of how this problem can be solved with and without machine learning, the power it can provide should be evident. But semi-supervised machine learning is just one of many concepts under the machine learning umbrella. In our next post, we’ll explore other models that help solve different problems.
In the meantime, if you’re interested in learning more, you can build you own intelligent email routing model with code we’ve made available in Github. You can download it here.By Saptarshi Mukherjee, Product Marketing Manager and Prashant Dhingra, Machine Learning Lead, Google Cloud
(function() {
function setup() {
var url = ‘https://polygraph-cool.github.io/google_cloud_1/index.html';
var pymParent = new pym.Parent(‘google-cloud-embed’, url, {});
}
function check() {
var el = document.querySelector(‘#google-cloud-embed’);
if (el) setup();
else setTimeout(check, 500);
}
check();
})()
Quelle: Google Cloud Platform
Published by