Artificial intelligence for the web
Richard Mattka introduces you to the powerful field of artificial intelligence, exploring how you can use it and create your own chatbot for your next web project
Artificial intelligence (AI) is an integral part of our world, embedded in nearly every technology we have. AI is in the Google searches we run, the voice commands we give Alexa and the map directions we follow. It’s part of ordering our morning coffee and in the navigation system in our cars. Our AI-powered phones, which are never out of reach, have become an extension of our physical selves and our very identity. AI has the potential to make almost everything we do easier and vastly improve our world.
As a technologist, it’s critical that you learn as much as you can about how to leverage these technologies and integrate them into your work.
So what is AI?
Artificial Intelligence (AI) is defined as a machine-based intelligence, as opposed to the biological-based intelligence of humans and animals. AI refers to machines performing functions of cognition, such as learning, planning and solving problems. But the definition seems almost too simplistic to capture the incredible range of incarnations of AI.
Communication, transportation, scientific discovery, medical research and service industries – all are enhanced
by AI. It performs a wide range of activities including game theory, electronic trading, robotic automation and exploring the vastness of space.
Another way to define AI is as intelligent ‘agents’, which can perceive their environment and take actions towards achieving their goals. You’re going to learn how to create your own intelligent agent later on in this article, in the form of a chatbot.
Blurred lines and the challenge of defining AI
Defining AI has become increasingly difficult because technology evolves so rapidly. We tend to extend the definitions of AI as tasks performed by AI become routine. Basic tasks such as autocorrect or autocomplete hardly seem notable today, in the face of self-driving cars and computer vision.
In fact, AI is so integrated with our everyday experience we may be hardly aware of it. We may lose sight of where we end and AI begins. AI is so prevalent it is becoming invisible to our perception. Instant search, with most relevant results at out fingertips, is just expected. Massive collective knowledge available with a voice command. Your phone shows you directions to a location that you are ‘most likely’ to be going to next (yep, your phone knows you walk to the coffee shop every morning before work).
Disciplines of AI
Despite the ever-changing definitions, there are several identifiable objectives or disciplines within AI. Some applications are but are not limited to: Knowledge reasoning Machine learning Natural language processing Computer vision Speech recognition Robotics Virtual reality Data mining Game theory
AI knowledge reasoning
Knowledge reasoning is defining information in a format that a computer system can use to solve complex problems such as diagnosing a medical condition or having a dialogue using natural language. It combines problemsolving psychology and logic to automate reason.
Machine learning uses statistical techniques to ‘learn’ without being explicitly programmed. Using data samples, the AI progressively improves by analysing them and making continual predictions. Some examples include Amazon recommendations, Siri voice recognition, spam filtering and computer vision.
Natural language processing
Natural language processing (NLP) focuses on the interactions between machines and human languages. It is the objective of NLP to process and analyse vast amounts of natural language data, to have improved ‘natural’ communication between humans and machines. This field of AI includes speech recognition, understanding language and generating natural language.
Computer vision is an incredible field that focuses on how AI can gain comprehension or understanding from digital images or videos. The objective is to automate what biological visual systems can do and make AI see and understand what it is looking at. Examples include detecting events, tagging and classifying images, motion tracking in videos, image or scene restoration and object recognition.
AI in web applications
Websites and apps can have a variety of moving parts, including front-end creative, server-side processing, APIs, data storage and various forms of interconnectedness. AI can plug in any of these components. On the front end, you can connect voice commands, chatbot interfaces or reactive WebGL creative elements. On the back end, databases use intelligent algorithms to maximise speed and analysis. APIs can provide a layer of abstraction from a wide range of AI functions, from predictions to collective training. On the front-end, you can connect voice commands, chatbot interfaces or reactive WebGL creative elements