Name: Introduction to Data Analytics and AI
Uploaded: 2024-10-05
Duration: 4 min 14 s
Description: Introduction to Data Analytics and AI. © Copyright New Zealand Skills and Education College 2024.

Introduction to Data Analytics and AI. © Copyright New Zealand Skills and Education College 2024. 

Scene 1

Artificial Intelligence (AI). 6/10/2024. A blue circle with white text Description automatically generated. 

Scene 2

A blue circle with white text

Description automatically generated

AI is a discipline AI simulates human intelligence in machines that are programmed to think, learn, and perform tasks autonomously.  The goal of AI is to create systems that can mimic human cognitive functions such as learning, problem-solving, perception, and language understanding.

Machine Learning (ML). ML is a subset of Artificial Intelligence (AI) It is a method of data analysis ML uses algorithms and statistical models that allow machines to identify patterns and make decisions based on data. In essence, machine learning is about teaching computers to learn from data and perform tasks such as prediction, classification, or clustering, without human intervention.. 

Scene 3

ML is a subset of Artificial Intelligence (AI)  It is a method of data analysis ML uses algorithms and statistical models that allow machines to identify patterns and make decisions based on data. In essence, machine learning is about teaching computers to learn from data and perform tasks such as prediction, classification, or clustering, without human intervention.

Artificial Intelligence
Computer programs obtaining 8
detecting information, reacting to
and learning from it
Machine Learning
Continuously evolving algorithm
that improves based on the quantity
of data available

Supervised Learning. 6/10/2024. © Copyright New Zealand Skills and Education College 2024. 

Scene 4

The model is trained on a labeled dataset. input data and the corresponding output are provided.

Statistics 

 Regression Models
 Linear Regression: Models the relationship between the pollutant concentrations and the AQI value using a linear equation. Decision Trees: Splits the data based on feature values to predict continuous AQI scores. Random Forest Regressor: An ensemble of multiple decision trees used to reduce overfitting and improve predictive accuracy. Support Vector Regressor (SVR): Models nonlinear relationships between pollutants and AQI. Neural Networks: Can capture complex patterns in data, making it effective for predicting AQI in urban environments with multiple pollutants.

Table 

 Classification Models for AQI Levels
 Logistic Regression: Classifies AQI into different levels (e.g., Good, Unhealthy) based on feature values. k-Nearest Neighbors (k-NN): Assigns AQI levels based on similar past observations. Support Vector Machines (SVM): Creates boundaries to separate different AQI categories. Random Forest Classifier: An ensemble of decision trees that improves classification accuracy for AQI levels.

Artificial Intelligence
progranB obtaming B
•a ming it
Machine Learning
qt_mtitv
ot

Model-Free RL: Learns directly from interactions without an explicit model of the environment. Examples: Q-Learning, SARSA. Model-Based RL: Uses a model of the environment to predict future states and plan actions. Examples: Monte Carlo Methods, Dynamic Programming. Policy-Based Methods: Focuses on learning the policy directly. Examples: REINFORCE, Actor-Critic Methods. Value-Based Methods: Estimates the value of actions to derive the optimal policy. Examples: Q-Learning, Deep Q-Network (DQN). Deep Reinforcement Learning: Combines RL with Deep Learning to solve complex, high-dimensional problems. Examples: Deep Q-Networks (DQN), Proximal Policy Optimization (PPO), Deep Deterministic Policy Gradient (DDPG).. 

Scene 5

Learns to make decisions by interacting with its environment.  Focuses on training agents through the concept of rewards and penalties to achieve a goal.  The primary objective is to learn a policy that maximizes cumulative rewards over time.

Multi-Agent Deep Reinforcement Learning for Connected Autonomous Driving

Model-Free RL:

 Learns directly from interactions without an explicit model of the environment.

 Examples: Q-Learning, SARSA.

 Model-Based RL:

 Uses a model of the environment to predict future states and plan actions.

 Examples: Monte Carlo Methods, Dynamic Programming.

 Policy-Based Methods:

 Focuses on learning the policy directly.

 Examples: REINFORCE, Actor-Critic Methods.

 Value-Based Methods:

 Estimates the value of actions to derive the optimal policy.

 Examples: Q-Learning, Deep Q-Network (DQN).

 Deep Reinforcement Learning:

 Combines RL with Deep Learning to solve complex, high-dimensional problems.

 Examples: Deep Q-Networks (DQN), Proximal Policy Optimization (PPO), Deep Deterministic Policy Gradient (DDPG).

Recommendation Algorithms. machine learning methods designed to suggest relevant items to users based on their preferences, behaviors, and interactions. e-commerce online streaming services social networks and content platforms. Goal: to provide personalized experiences by predicting what a user might like or need. Why Use Recommendation Systems? Personalization: Tailoring the content for each user to enhance engagement. Increased Sales/Engagement: Recommending items users are likely to buy or interact with. Content Discovery: Helping users find new content or products they might not have discovered otherwise. Improved User Experience: Streamlining decision-making for users by reducing information overload.. 

Scene 6

machine learning methods designed to suggest relevant items to users based on their preferences, behaviors, and interactions.  e-commerce online streaming services social networks and content platforms.  Goal: to provide personalized experiences by predicting what a user might like or need. Why Use Recommendation Systems? Personalization: Tailoring the content for each user to enhance engagement. Increased Sales/Engagement: Recommending items users are likely to buy or interact with. Content Discovery: Helping users find new content or products they might not have discovered otherwise. Improved User Experience: Streamlining decision-making for users by reducing information overload.

Deep Learning. Transformed AI by enabling machines to solve highly complex problems and achieve performance levels comparable to humans in specific tasks. Its growing popularity continues to push the boundaries of what AI systems can accomplish.. 

Scene 7

Transformed AI by enabling machines to solve highly complex problems and achieve performance levels comparable to humans in specific tasks.  Its growing popularity continues to push the boundaries of what AI systems can accomplish.

Artificial Intelligence
Computer programs obtaining 8
detecting information, reacting to
and learning from it
Machine Learning
Continuously evolving algorithm
that improves based on the quantity
of data available
Deep Learning
A set of techniques using
convolutional neural networks
trained on very large
amounts of data

MACHINE LEARNING
inputs
inputs
feature extraction
learning
DEEP LEARNING
feature extraction • learning
Outputs
outputs

Model=Training(Data+ Algorithm). Validation: tune model parameters and evaluate its performance during training. Test: assess the final performance of the trained model.. 

Scene 8

*rvcnn
Discw•y
Elie•øan
Msualtmlon
lusterin
imensionar
Reduction
nsupervise
Learning
Cwmmu•
lassificatio
Supervised
Learning
Cu•ton•••
At—dslrg Popuhrlty
PredBon
egressio
Machine
Pmdctlgn
Learning
einforceme
Learning
Game
SKII *'Wismon

Validation: tune model parameters and evaluate its performance during training. Test: assess the final performance of the trained model.

Artificial Intelligence
Machine Learning
that imgmes the
of
p Learning
A ot
of data

Scene 9

A rapidly growing sub-field—has received the most significant investment since ChatGPT's launch in late 2022.
  Unlike traditional AI, which typically classifies or analyzes existing data, generative AI focuses on creating new data that mimics human creativity.
 OpenAI’s GPT-3 and GPT-4: Large language models that generate human-like text for chatbots, content creation, and more. DALL-E: Generates images from text descriptions. StyleGAN: Creates highly realistic images of faces, objects, and scenes. DeepDream: Generates psychedelic and artistic interpretations of existing images.

Warning with solid fill  Ethical and Legal Issues