Page 117 - Dig_CodeAI_V2.1_Class_8
P. 117
Data Science
Data science in AI refers to the process of collecting, analysing, and interpreting large sets of data
to enable AI systems to learn, make predictions, and improve their performance. It involves using
various techniques, such as statistics, mathematics, and computer science, to extract meaningful
insights from data.
In AI, data science is essential because AI systems need data to learn. For example, if you want an
AI to recognise images of cats, you need to provide it with a lot of pictures of cats. The AI uses data
science techniques to analyse these images, learn the patterns that define a cat, and then use that
knowledge to identify cats in new images.
The human eye is estimated at
576 megapixels with complex
processing, while AI machine
vision cameras can reach up
to 100 megapixels.
Developed by scientists at DAMO Academy, Alibaba's global research program, the Alibaba
model scored 0.54 in the MS Marco question-answering task, surpassing the human score of
0.539, which evaluates a machine's ability to use natural language.
Computer Vision (CV)
Computer Vision is a very popular field of AI that trains a computer to understand and interpret the
visual world. Human vision starts at the “eyes” but machine uses digital images from a camera for
vision. Deep learning models and machines accurately identify and classify objects that act according
to what they see, using digital images from camera.
According to Fei-Fei Li, computer vision is defined as “a subset of mainstream artificial intelligence
that deals with the science of making computers or machines visually enabled, i.e., they can analyse
and understand an image.”
Applications of Computer Vision
From recognising to processing the live action of a football game, computer vision challenges and
surpasses human visual abilities in many areas.
AI Domains 115

