PolyU EIE1005 e-Learning 1 - Introduction to Artificial Intelligence and Data Analytics

What is Data

• Definition: Individual units of information.
• Representation: Usually by variables.
• Source:
  • Scientific Research
  • Business Management
  • Government

What is Data Analytics

• Definition: The process of inspecting, cleaning, transforming, and modelling data.
• Goal: To discover useful information, inform conclusions, and support decision-making.
• Characteristics:
  • Input: Data-driven; the more data, the better—improves reliability and credibility.
  • Methods: Interdisciplinary approaches combining mathematics and computer science.
  • Output: Discovery of knowledge or actionable information from data.

The History of Data Analytics

• Data analytics saves people’s lives:
  Example: Cholera outbreak investigation.
• Big data challenge:
  U.S. Census → Deployment of Hollerith Tabulating Machine.
• Jump leap:
  • Manhattan Project: Catalyst for “big science”.
  • Space Program.
• The Era of Big Data:
  • Better models:
    Rule-based → Statistical → Deep Learning
    (Increasing number of variables to fit data more accurately)
  • Better computing resources:
    More powerful RAM, CPU, GPU, etc.
  • More data.

Characteristics of Big Data: 4V

• Volume（体量）: Enormous amounts of data to process.
• Velocity（速度）:
  Batch data, real-time data, streaming data—response times from milliseconds to seconds.
• Variety（多样性）:
  Structured, semi-structured, unstructured (e.g., text, images, multimedia).
• Veracity（真实性）:
  Data consistency, completeness, and cleanliness.

Applications of Data Analytics

• Product Recommendation:
  Recommend items to customer X based on items previously rated highly by X.
• Ranking Pages:
  Compute the importance of web pages using link graph analysis (e.g., PageRank).
• Artificial Intelligence:
  (Application area leveraging data analytics for intelligent systems.)