OER·harvester

← Back to the library
Zenodo PDF resource

Foundations of Artificial Intelligence & Machine Learning

Licence
OPEN CC-BY-4.0
Authors
Nidhi Sharma, Honey Singh, Ajay Sharma, Deepak Dagar
Published
2026-07-28 · Zenodo
Language
eng
Length
37166 words
Type
narrative text
Open ↗ Download Open original ↗
Understanding Probability

Probability measures the likelihood that an event will occur. It is generally represented as a value between 0 and 1.

  • A probability of 0 means the event is impossible.

  • A probability of 1 means the event is certain. For example:

  • the probability of obtaining heads from a fair coin toss is 0.5

  • the probability of rolling a six on a fair die is 1/6 Probability helps Machine Learning systems estimate uncertain outcomes using available information. In AI systems, probabilities are continuously calculated while:

  • predicting user behavior

  • recognizing speech

  • detecting diseases

  • identifying objects

  • recommending products Machine Learning models rarely make decisions with complete certainty. Instead, they estimate the most likely outcomes based on data patterns.

Figure 3.4: Basic Probability Representation

The figure illustrates the concept of probability as a numerical measure representing the likelihood of events ranging from impossible outcomes to certain outcomes. Random Variables A random variable represents a measurable quantity whose value depends on uncertain outcomes. For example:

  • weather conditions
  • stock market prices
  • customer purchases
  • examination scores can all behave unpredictably and are therefore treated as random variables. In Machine Learning, random variables are used extensively because real-world systems involve uncertain and changing information.

For instance, a predictive healthcare model may estimate the probability of a disease based on symptoms, medical history, and test results. Since future outcomes cannot be guaranteed, the system relies on probabilistic analysis. Probability Distributions A probability distribution describes how probabilities are assigned across different possible outcomes. Different Machine Learning problems involve different types of probability distributions. Some common distributions include:

  • normal distribution

  • uniform distribution

  • binomial distribution Among these, the normal distribution is especially important in statistics and Machine Learning. Normal Distribution The normal distribution, also called the Gaussian distribution, is one of the most widely used probability distributions in data science. It is represented by a bell-shaped curve where:

  • most values cluster around the mean

  • extreme values occur less frequently Many natural and human-generated datasets approximately follow normal distributions.

Examples include:

  • student examination marks
  • human height
  • measurement errors
  • product demand patterns Machine Learning algorithms often assume normal distribution properties during analysis and prediction.

Figure 3.5: Normal Distribution Curve

The figure represents the mathematical form of the normal distribution. The bell-shaped structure demonstrates how values are concentrated around the mean while extreme values occur less frequently.