My Life Blog

Define artificial intelligence

Artificial intelligence (AI) is gaining traction as a exponential rate, where it was once in special use cases, not almost any and everyone has the capabilities to use AI. To some AI is somewhat of a mythical or abstract idea, with little to no fundamental understanding of concepts.

When you tell people that we have been using AI in the consumer market for years now, most are surprised. In the world of today (2024), we use AI every day. With such uses as:

• digital assistants
• cellphone applications
• automotive technologies
• productivity tools

So, what exactly IS artificial intelligence? According to Microsoft, “… we tend to think of AI as software that exhibits one or more human-like capabilities.”

Some of the possible capabilities in use today are:

• visual interpretations
• text analysis and conversations
• speech recognition
• decision making

Understand AI

The concepts of AI can become entangled with many confusing topics and terminologies. This is the point that some individuals tend to abandon ship and begin to “get lost in the sauce.” There are three most important terms to understand when thinking of AI:

1. Data science
2. Machine learning
3. Artificial intelligence

Data Science

This is the key fundamental requirement for anything when it comes to AI. This discipline is where all the important and pertinent information derives from and how we are capable of creating these intelligent systems. Data science is used to process and analyze data, primarily with the use of statistical techniques to help recognize relationships and patterns within any dataset. These statistical techniques also begin the process of defining models to assist in traversal of noticed patterns.

Machine Learning

Diving a layer deeper into data science we have a subset which is called machine learning, this concept deals with training and validating models, commonly referred to as predictive models. Data scientists will gather and combine all the data and use the dataset to train models with the help of algorithms to make us aware of relationships between the features in the data. These datapoints (features) are then used to predict values which are labels.

Artificial Intelligence

Building off of machine learning and its concepts we use the found features and labels and create software to mimic human characteristics of intelligence. This process can be several different methodologies, such as supervised or unsupervised learning. Keep this in mind that AI is only the continuation or culmination of both data science as the foundation and machine learning for pattern recognition.