Feature Friday – Machine Learning
Behind the scenes we are working hard on our latest product version, V4. To keep you informed of all developments, we will share a new or updated feature every Friday in our “Feature Friday.” This Friday: Machine Learning.
What is Machine Learning?
Nowadays, a recurrent “hot topic” in the professional world, and not only in software domain, is Machine Learning (ML). When talking about Machine Learning most people confuse this with Artificial Intelligence (AI), but ML is only one of the multiple broad research fields within AI, with a strong focus on software development and a great potential on data driven companies.
But, what is Machine Learning exactly? ML is an AI discipline in which computers are taught and trained on how to resolve given practical tasks from huge amounts of data, too big or too complex to be processed or hardcoded by humans. By using data, computers gain new insights, being able to recognize patterns within them. If you combine your data and the use of ML mathematical algorithms, you can build models that can solve complex problems and make predictions more efficiently. We can use these results to improve and simplify our workflows.
In short, we can see ML as a human brain: the more data it receives to analyse, the more patterns it will recognize and the model will be more general, resulting in more accurate predictions.
How does Machine Learning work?
An example: let's say we would like to train a computer to recognize the difference between a picture of a car and a bike. We can process some image data that we have already classified as car or bikes, feed them to the computer, and it will be in charge of seeking these statistical patterns that will lead to an accurate model, able to classify new images. The more data it gets, the more it will be able to build models to recognize the difference in the future, on its own
ML is not a new research field, it was born in the 60s and it has been developing since then. Nowadays it is becoming a mainstream technology used worldwide, and we might not be aware of it. What do you think of facial and text recognition or spam filters in your inbox? It's all the work of ML.
How SmartLockr uses Machine Learning
In SmartLockr we will use ML to reinforce the security on your email system, by supporting our clients in the avoidance of data leaks and unsafe emails sending. We use ML technologies to process client’s emails through text recognition, being able to detect the degree of confidentiality of the email and let them know in case there might be some risks on it.
Also, we will use it to create more awereness among email users. Are you about to send confidential data? ML will recognize this content as confidential and will give you a notification. By doing so, you will always be aware of what you are sending out.
We are aware of the confidentiality of our clients, so the system will be built and implemented regarding to the new European General Data Protection Regulation (GDPR). The algorithm will work only with non-confidential data and clients email content in their own company environment which means that all the data will be used in a secured environment.
Being able to use Machine Learning, increases the user friendliness of SmartLockr. But, this is only one of the many new features SmartLockr has to offer. If you would like to learn more about our product, you can read about our Secure Emailing Solution or perhaps you want to Try Our Product?