Data Analytics Course Overview

Class Rules

• Students can do anything except make noises (e.g., chatting, singing).
• Students can interrupt with questions.
• Attendance is required according to university policy.
• Students need 80% attendance to sit the final exam.

Course Assessment

• Final exam is 50%.
• Assignments are 20% (individual).
• Projects are 30% (2-3 people per group, requiring a report and presentation).
• Cheating and plagiarism will result in no marks.

Additional Suggestions

• Final grades are based on points, not the accumulation of grades.
• Students start with zero points and earn points throughout the class.
• Communication is key—contact the instructor if there are problems or issues.
• Inform the instructor before deadlines if you are unable to meet deadlines for quizzes or assignments.

Data Challenges

• Data is everywhere; you have data, you have everything.

Analytics

• Key Concepts: reinforcement learning, predictive analytics, image recognition, gradient boosting, neural networks, optimization, supervised learning, unsupervised learning, edge analytics, deep learning, data analysis, transfer learning, sentiment analysis, gradient boosting, support vector machine
• Why Analytics?: (this is likely a question, not a concept) The significance of analytics.
• Levels of Analytics: descriptive (what happened), diagnostic (why did it happen), predictive (what will happen next), prescriptive (what should we do about it), planning (what is our plan).

Career Choice

• Careers in analytics are highly sought after.
• Acquiring analytical skills can help differentiate from other applicants.

Artificial Intelligence Eras

• Eras include computer vision, machine learning, natural language processing, forecasting and optimization.

Machine Learning

• Machine learning learns from past data.
• Humans learn from experience.

Machine Learning Classification

• Concepts mentioned include recommended systems, targetted marketing, customer segmentation, image classification, feature elicitation, fraud detection, customer retention. Methods include unsupervised learning, reinforced learning, supervised learning, regression, diagnostics, forecasting, predictions.

Clustering and Dimensionality Reduction

• Clustering: grouping similar data points together.
• Dimensionality Reduction: reducing the number of variables in a data set.

Graph Analysis

• Analysis of relationships between data points.
• Likely involves various visual representations.

Natural Language Processing (NLP)

• Types of NLP: Information Retrieval, Machine Translation, Sentiment Analysis, Information Extraction, Question Answering, Chatbots

Computer Vision

• Object Classification: identifying objects in images.
• Object Identification: recognizing specific objects.
• Scene Reconstruction: recreating scenes from images.
• Motion Analysis: tracking movement within footage.
• Classification and Localization: Identifying single objects in images and their location within the frame (e.g., CAT).
• Object Detection: Identifying multiple objects and their location in images (e.g., CAT, DOG, DUCK).
• Instance Segmentation: Identifying and outlining specific individual instances of objects in images (e.g., multiple cats).
• Human Visual: likely a comparison of human visual perception to that of computer vision.
• Light Rays, Retinas and Processing: describing light and visual processes in humans and computers.

Goals of Computer Vision

• Goal: to compute 3-D shapes of the world, recognize objects and people, improve photos (computational photography), and forensics.

Digital Image Processing

• Image transformation
• Image compression
• Image restoration
• Image enhancement

Explore the essential rules and assessment criteria for the Data Analytics course. Understand the importance of communication and project collaboration, as well as the key concepts such as reinforcement learning and predictive analytics. This quiz will help reinforce your knowledge of course expectations and data challenges.