Quick Intro to Machine Learning for Non-Tech People
Machine Learning has gradually spread into our lives in different ways. For example, your newly recommended music from Spotify and videos from Netflix. People are talking about machine learning all the time and you might have already hear about it several times in TV shows, news, or even in animations. It seems to be a panacea for lots of hard problems (Of course, it is not 😉)
However you might still be confused about what it really is? Here is a simple quick intro to Machine Learning for non-tech people. Hope it can help you get a basic idea about what it is and to set a reasonable expectation on machine learning about what it can help us to achieve nowadays.
What is Machine Learning
Machine learning is a field of computer science that gives computers the ability to learn without being explicitly programmed. Then they can apply what they have learnt from existing data to forecast future behaviors, outcomes, and trends to solve problems. For example, an algorithm can be trained with cat photos to recognize cats; the same algorithm can also be trained with bicycle photos to recognize bicycles without changing a line of code.
Two Main Categories
There are two main categories — supervised learning and unsupervised learning. The majority of machine learning’s practical value today is from supervised learning. The difference between supervised learning and unsupervised learning is simple. For supervised learnings, you have input variables (X) and an output variable (Y) and you use an algorithm to learn the mapping function from the input to the output to get Y = f(X). You are trying to learn about the relationship among data. For unsupervised learning, you only have input data (X) and no corresponding output variables. The goal is to model the underlying structure or distribution in the data in order to learn more about the data. Unsupervised learning is like you give the machine just some data and tell it “I don’t know what is happening over there, just try to figure something out~ good luck!”
Machine Learning Process
Simplified, a machine learning project usually includes 4 main steps: preparation, training, testing/evaluating, and deployment. Let’s make a metaphor here to make it easier: You need someone to do a task for you. However, the only workforce you have right now is some people who know nothing. To accomplish the task, you need to send these people to school to get trained. Then, after graduation, they will be capable to take the task.
Let’s talk about the steps one by one.
1- Preparation
Prepare the Textbook and Pick the People You Want to Send to School
There is a reason why you want to user machine learning. To start a machine learning project, you need to be clear about this reason: what is the goal you want to achieve and the problem you are trying to solve. Then, you can do some preparation works including preparing data and selecting appropriate algorithm.
Prepare Data
Preparing the dataset ready for training is just like preparing the textbook. There is so much knowledge that you can let your people learn. You must have the knowledge be clearly organized and presented in an understandable way in the textbook before they can really learn it. Knowledge in the real world can be scattered and unorganized. Similarly, raw data is often noisy and unreliable and may have lots of missing values. Using such data will product misleading results. To get the dataset be machine learning ready, you need to do some pre-processing and cleaning works on the datasets.
Choose Algorithm
Use which algorithm depends on the dataset that you want to train on; how much time and resources do you have; what do you want to use the algorithm for, etc. Each algorithm has its own talented field, some are good at dealing with images, some are gifted in processing natural languages… There are numerous algorithms in the world and new ones are born frequently. You need to identify the most appropriate ones based on different considerations.
2- Training
The training process of a supervised learning is the process of the algorithm learning from the data. Depends on what problem you want to solve and which dataset and algorithm you have chosen, it can be a process for the algorithms to learn how to do things as simple as recognize a cat, decide whether a sentence is positive or negative, to as complex as decide the fraud risk of a loan, etc.
3- Testing/Evaluating
Now you have your algorithm trained. But you do not know whether it has really learnt what you want them to do. So what will you do? In school, students will take exams. Similarly, in a machine learning project, trained algorithms need to be tested/evaluated by using some test datasets. If the test result is not satisfied, just like a student failed an exam, the training will need to be conducted again.
4- Deployment
Now, after all the hard work of preparation, training, and testing, your algorithm is ready to take tasks and create values for you! It can be deployed as a web services, put into an app, etc. Just like any other non-machine learning projects, after we have gotten the core part work, there will still be lots of hard work needed in order to make it really production level ready :- )
Of course, doing a machine learning project in real life is much more complex than this simple four steps intro. There are numerous considerations need to be taken care of and various challenges need to be overcome. Here are some readings to help you to know more about machine learning:
Book Draft by Andrew Ng: Machine Learning Yearning teach you how to make the numerous decisions needed with organizing a machine learning project
Documentation from Microsoft Azure What is Machine Learning A good documentation which can help you learn about some machine learning terminologies.
Blog on Medium The Non-Technical Guide to Machine Learning & Artificial Intelligence which includes lots of gateways for you to explore more about Machine Learning and Artificial Intelligence by yourself.
If you have any feedbacks, please leave a comment~ :- ) I will be very grateful if you could help with me be writing better blogs. Thank you~
Some of the Other References:
This story is published in The Startup, Medium’s largest entrepreneurship publication followed by 295,232+ people.
Subscribe to receive our top stories here.
