Relationship between machine learning and deep learning

“It refers to an attempt to artificially realize the same intelligence as humans on a computer, or a series of basic technologies to achieve it,” but researchers have different opinions and don’t have a strict definition.

Artificial intelligence is generally interpreted as “the concept and technology of artificially imitating human intelligence.”

AI (Artificial Intelligence) is required to enable a computer to “learn” and “guess” based on the knowledge like a human. This is why AI uses complex platforms and algorithms. AI is used in various places, as familiar examples, speech recognition of smartphones, self-driving cars that avoid obstacles, image retrieval or web page search on the Internet, robot control in the industrial field, and image processing.


Neural networks are a mathematical representation of neurons and their connections in the human brain, called artificial neurons. In recent years, Artificial Intelligence (AI) has become a fad. Neural networks are a basic mechanism that you should know when learning machine learning, deep learning and so on.


AI (Artificial Intelligence) is a comprehensive concept and technology, and “Deep Learning” is one of the methods that support AI.

As an introduction to learning neural networks, you need to understand the mechanisms, structures, machine learning, relationships with deep learning, specific examples of various approaches, and the flow of deep learning.

Each artificial neuron has a simple mechanism. However, a combination of such a large number of functions can result in a complex function approximation. This is the big feature of neural networks.

There are often cases where the traditional machine learning method does not work when sorting or regression cannot be done unless the complex function approximation is done. There are more cases in which deep learning techniques are used to solve such problems.

Using deep learning techniques, there are cases in which recognition accuracy is improved dramatically compared to the conventional method. Therefore, deep learning is gaining attention in the world. Recently, it has been used in a wide range of fields such as the field of self-driving.


Deep learning is a machine learning method by a multilayered (more than four layers in a narrow sense) neural network (deep neural network; DNN).

Before deep learning, more than four layers of deep neural networks were unable to be fully learned due to technical problems, and their performance was poor. In recent years, however, it has become possible to learn them well by studying the learning of multi-layered neural networks, improving the skills of computers required for learning, and facilitating training data procurement by the development of the web.

As a result, the company showed high performance that overcomes the other methods in order to solve the problems of voice, image and natural language, and it was popularized in the 2010’s.


Big data reminds you of a lot of data, but it doesn’t only mean it.

Big data refers to different types of data with different natures and forms.

Actually, big data consists of three types of data: volume, variety, and frequency of data generation and update velocity, and all of which are important.