Artificial intelligence, machine learning and deep learning

# Artificial intelligence, machine learning and deep learning **Introduction** The last decade has seen artificial intelligence (AI) deliver impressive results in areas ranging from image detection to self-driving cars, and most people are familiar with AI performing tasks on our behalf like powering online customer support as seen on Zendesk, Amazon Alexa or even recommending products as we do so on e-commerce platforms like Shopify. As we go deeper into the AI journey, there are bound to be more AI-based applications making their way into every aspect of our lives. From improving medical and medical research, powering customer service chats to answering security questions, these AI platforms are going to continue enabling us to get more out of our working days while freeing us from menial tasks that distract us from the bigger picture. It wasn’t long ago that computers on our desktops and in our pockets could merely process ones and zeroes. Currently, computers can recognize images and detect motion. They can understand natural language, detect our emotions and face, as well as make sophisticated predictions about the world around them. This has taken AI far beyond what people once thought was possible and set the world on a path to realize the possibilities of man(machine)kind. “Artificial Intelligence”, “Machine Learning”, and “[Deep Learning](https://www.ibm.com/cloud/learn/deep-learning)”. You often see them used interchangeably in media outlets, but what do these terms actually mean? Are the three really the same, in essence? This article will attempt to explain how each of these terms is connected — more precisely what is the difference between them — and how each of them has been instrumental in advancing recent achievements in AI development. Artificial intelligence (AI) is commonly defined as the ability of an artificial system to perceive its environment and take actions that maximize its chance of success at some goal. As such, AI is often equated with machine learning, decision-making and planning. **Artificial Intelligence** Artificial intelligence was first introduced as a term in 1955 by John McCarthy as part of preparations for the [Dartmouth conference](https://en.wikipedia.org/wiki/Dartmouth_Conference) that was held in 1956. The participants of conference described artificial intelligence as: “The study is to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it.” Artificial intelligence is subdivided into two areas: strong, general AI, artificial general intelligence (AGI) and weak, narrow AI. Strong AI is as advanced computers and other machines with human-level intelligence. Strong AI does not actually exist at the moment, but various companies are working hard on developing it. With current AI, we are able to perform various tasks, like translating text from one language to another. However, not all of us can do this or other things like driving cars or performing specific surgeries. Strong AI tells about the development of machines that could do these tasks at a level comparable or better than humans. Such AI does not exist at the moment but it is a concept many of world's leaders are investing in to grow businesses and economies. We still live in a narrow AI world and we will for some time until we see larger advances in AGI. **Machine Learning** Machine Learning is a subset of artificial intelligence concerned with the creation and development of systems that can learn from data. From self-driving cars to smart assistants, machine learning is being used in many aspects of our lives. Our company works in AI field and has developed many platforms that utilize the machine learning models. As part of our platform BittsAnalytics we collect 1+ million tweets each day and using SVM machine learning model compute their sentiment. From these we then compute [cryptocurrency fear and greed](https://bittsanalytics.com/crypto-fear-greed-index/) as an indicator for overall market. Cryptocurrency fear and greed gives an insight in current emotions in the crypto markets. Crypto market has risen a lot since the invention of Bitcoin and the general crypto market cap is in trillions and now comparable to some of the tech companies. Machine learning algorithms are often divided in three groups: supervised machine learning unsupervised machine learning reinforcement learning Most of our models that we developed have been in the supervised machine learning field. A great library for this is sklearn: https://scikit-learn.org/stable/ We have developed a ML model which tries to predict which niche from a list of 10k niches can be considered as [the best seo niche](https://bestseoniches.com/). We use various features for this: - strength of optimization - domain age - content length - content variation (number of distinct words) - link strength (we use open page rank for this) - volume domination by top domains in the niche Our another Saas and [AI](https://www.sas.com/en_us/insights/analytics/what-is-artificial-intelligence.html) platform - UnicornSEO is specializing on another important part of content marketing - determining best [seo keywords for your niche](https://www.unicornseo.com). It does this by computing Z-scores of individual factors and from these the weighted average total score of each keyword in the niche.