Guide to explain Machine Learning to your GrandMa

Two simple words

Machine Learning, aren’t new words but maybe the combination of them generates doubts and new questions about what this stands for, and some of those could be:

• Learning from machines?
• Learning with the help of computers?
• Learning through online courses?

This could be the first approach to the concept and a start point to say what is not Machine Learning, and that helps a lot in the purpose to explain somebody a new topic.

But perhaps there are other sides to begin this almost non-technical explanation, because, in the past, science fiction movies and even radio series propose the idea that machines at some point in the future can learn by themselves, surpass their creators and dominate the world, obviously with terrible consequences, so the concept of Machine Learning isn’t new, not clear, but not new, then we have some ideas to clarify and many words to put in the correct order to explain this new term to our adorable grannies.

How to Learn

The most common task for human beings has been cooking, and the development of recipes within families even through generations is well known, and we can ask our grandmother how was her process to be the really good cook she is, and we’ll gather the correct information, she’s going to say she learned from his mother, at first only looking at her and later trying to repeat the process and improved it depending on the taste of the food, or on the comments of the rest of the family, and that story is a gold mine to achieve our goal.

What is Machine Learning?

Now we need to look right directly to the eyes of that beloved being and say:

“Machine Learning is the science of getting computers to learn and act like humans do, and improve their learning over time in autonomous fashion, by feeding them data and information in the form of observations and real-world interactions.” 1

It seems like too much, but if we go backward we can resume previous ideas and put them together to enlighten the bomb we launched before.

Grandma knew that one-day machines could learn, so you can tell her today is the day and that’s happening right now, but not in the violent way she thought, actually they are learning the same way she learned to cook, by repeating processes and collecting information, the difference with today’s machines is that they can collect huge amounts of data and are able to execute tasks at high speed so they can learn really fast.

Now she can say, machines are supposed to do what humans order, and the answer to that is yes they are, but now we decide that we are not going to say them step by step what they are going to do, instead, we choose the most expensive and difficult tasks and show them the inputs and the results and give them the tools to learn and improve.

Oh-oh giving tools to a computer, it sounds weird, OK it’s time to go deeper. What about statistics, is this a new word? nop, grannie knows about them because she had voted in presidential elections and she has listened about the probability or possibility that a candidate has to win, and it is a good way to introduce her to the world of statistics.

Once again, one bomb more:

“Statistics is the science of gathering and analyzing numerical data in large quantities. It is the science of learning from data, and measuring, controlling and communicating uncertainty, especially for the purpose of inferring proportions in a whole from those in a representative sample”. 2

That can be said like, the same way news channels interview people to know their preferences for a certain candidate, scientists, and many other people are doing the same in other fields to have enough information to give it to a computer to show it what happens when a task or event is done and then with that input force the machine to do the same thing to achieve the same results, like a human.

What are the applications of Machine Learning?

This conversation seems like a lack of purpose, why people need a computer to think and act like them? What are the applications of these?

Grannie has gone to the bank, and to the supermarket and to the doctor, so she can understand the next examples of the use of Machine learning.

Banking & Financial services: Machine Learning can be used to predict customers who are likely to default from paying loans or credit card bills. This is of paramount importance as machine learning would help the banks to identify the customers who can be granted loans and credit cards.

Healthcare: It is used to diagnose deadly diseases (e.g. cancer) based on the symptoms of patients and tallying them with the past data of similar kind of patients.

Retail: It is used to identify products which sell more frequently (fast moving) and the slow moving products which help the retailers to decide what kind of products to introduce or remove from the shelf. Also, machine learning can be used to find which two / three or more products sell together. This is done to design customer loyalty initiatives which in turn helps the retailers to develop and maintain loyal customers.

These examples are just the tip of the iceberg. Machine learning has extensive applications practically in every domain. The examples included above are easy to understand and at least give a taste of the omnipotence of machine learning.

Machines are going to replace humans in all fields?

This question can appear because of the fear of lack of employment for the people and the fear of a crisis generated because of the improvement and development of new technologies.

We can tell her that man is superior in two essential points: creativity and flexibility, which means that a human is able to look at something from a new angle and to find a solution that does not have to correspond to a given pattern. The latter means, however, to be able to react quickly to new influences in terms of work technology. In addition, intrapersonal and interpersonal intelligence is guaranteed, which allows us to understand our counterpart and the respective conversation situation and to be able to react accordingly.

Both are important above all in areas that require a high level of expertise and personal interaction. For example, in market research, the demand for specialized niche knowledge is extremely high. This is partly because they are areas that are too complex for Machines, the amounts of data that would be important for a corresponding analysis, often not available in sufficient form and some areas are still completely unexplored so that only one expert reliable judge what the right decision is. Accordingly, Machine Learning can be supportive by providing information when enough data is to be analyzed, which is then interpreted by experts.

What will be the jobs for the people in the future?

Our dear grandma could be worried about the future of her family and it’s time to calm down and tell her that everything’s all right, because there is and there will be many opportunities to have great jobs, and some of them and their responsibilities are:

Machine learning engineer: It’s usually considered as the initial role among all machine learning jobs. Machine learning engineers create algorithms to help decipher meaningful patterns from massive amounts of data. These people also develop applications, which can perform common tasks done by humans in order to generate effective results without errors.

Ability of developing highly-scalable distributed systems and addressing different business challenges by implementing machine learning algorithms are two major job responsibilities of machine learning engineers.

Data scientist: Being one of the most in-demand machine learning jobs, data scientists are primarily involved in gathering data from different touchpoints, analyzing and interpreting it, drawing insights and inferences, and coming up with forward-looking solutions for business concerns.

Robust programming skills together with strong knowledge in statistics are two crucial criteria for this position. They’re also required to source massive datasets located in disparate places to uncover actionable insights and information on which business decisions can be made. This position also entails identifying problems and working to rectify them.

Data analyst: These people are expected to be familiar with data retrieval, data visualization, data warehousing, Hadoop-based analysis, and other business intelligence concepts.

A strong background in statistics, mathematics, machine learning, and programming is required to excel in this position. Core responsibilities of these persistent data miners include designing and deploying algorithms, triaging code issues, culling information and identifying risk, and data pruning, among others.

Business intelligence developer: Apart from applying skills on AI and machine learning, a business intelligence developer also holds strong business acumen.

He/she is responsible for crunching huge chunks of data in order to derive business insights, and works on to increase the profits of the business from an array of perspectives. From designing and maintaining data to optimizing processes and workflows, these people are responsible for a business’s growth.

Research scientist: This interdisciplinary role requires the person to go back and forth while working between projects related to machine learning and artificial intelligence. He/she should be involved in natural language processing, deep learning, reinforcement learning, and computer perception, among others. Some of the key skills required to become a research scientist include distributed computing, parallel computing, as well as computer architecture and algorithms.

Apart from the above, machine learning jobs include positions like data warehouse engineer, software architect, designer in human-centered machine learning, computational linguist etc.

With this information, I think it’s possible to explain the concept and erase the wrong ideas about Machine Learning, its applications and how new technologies are here to develop a new world and a new future instead of replacing people because in fact machines are doing things humans are not capable to achieve in a lifetime.