BTE -210 Chapter 5: Artificial Intelligence & Machine Learning

Machine Learning Systems

-Optical character recognition -Face recognition -Fraud detection -Customer Segmentation -Topic identification

Topic identification

Categorize news articles as to whether they are about politics, sports, entertainment, and so on.

Machine learning algorithms

How well the program is trained makes the difference in the outcome: -supervised -unsupervised -reinforcement

Subfields of AI

Machine Learning Natural Language Processing Robotics Vision

artificial intelligence

Subfield of Computer Science which deals with building intelligent machines, agents or systems, which simulate intelligent living beings (human) behavior. These systems should be able to solve problems and evolve by themselves. Creating an intelligent agent that perceives the environment and makes decisions to maximize chances of achieving its goal.

Natural Language Processing

a branch of artificial intelligence that helps computers understand, interpret and manipulate human language, like speech and text. -human computer interaction Allowing scientists to embed the technology into objects around us, whether or not they have a screen:

Machine Learning

an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. focuses on the development of computer programs that can access data and use it learn for themselves. enables analysis of massive quantities of data.

Expert Systems

are rule based systems which utilize a predefined data set created by 'subject matter experts

Supervised Machine Learning

can apply what has been learned in the past to new data using labeled and classified examples to predict future events. Starting from the analysis of a known training dataset, the learning algorithm produces an inferred function to make predictions about the output values.

customer segmentation

divides a market into categories that share similar attributes such as age, location, gender, habits, and so on

Unsupervised machine learning algorithms

it explores the data and can draw inferences from datasets to describe hidden structures from data that has not been classified nor labeled.

Human-Computer Interaction

shifts from command base, on-screen interaction to CONVERSATIONAL, giving computers the ability to interact with us in a more natural way.

Optical Character Recognition (OCR)

the capability of specialized software to interpret the actual letters and numbers on a page to create a digital document that can be edited, rather than a flat picture

Ultimate goal of AI

to mimic 'human intelligence'. Signs of intelligence such as: -Learning -Understanding from prior experience to respond quickly to new situations -Making sense of ambiguous or contradictory messages (e.g. sarcasm, jokes) -Ability to change behavior or response, with new knowledge

Fraud Detection

uncovering patterns consistent with criminal activity