Perspectives
Lecture Notes
-
[ML Lecture 03] Methodology: Optimisation and Gradient Descent 1. IntroductionRegression constitutes a fundamental approach to supervised machine learning, in which a predictive function is estimated from labelled training observations. Previous discussions introduced linear and polynomial regression and examined the relationship between model flexibility, interpretability, and generalisation.This article examines the methodological principles underlying mo.. -
[Machine Learning 02] Regression: How Do Models Learn and Generalise? Regression is a fundamental technique in supervised machine learning, primarily used to predict continuous numerical outcomes.This post presents a structured review of the principles underlying regression models, including model formulation, parameter estimation, and performance evaluation. It further examines three key properties of machine learning models: flexibility, interpretability, and ge..
-
[DM Lecture 02] Part 3 Similarity, Distance and Data Preparation Part1: Understanding Data: Types and CharacteristicsPart2: Data Quality and Statistical DescriptionPart 3: Similarity, Distance and Data Preparation Similarity and distance measures are fundamental concepts in data mining and machine learning. They help quantify how closely related two data objects are. These measures are widely used in clustering, classification, recommendation systems and inf.. -
[DM Lecture 02] Part 2 Data Quality and Statistical Description Part1: Understanding Data: Types and CharacteristicsPart 2: Data Quality and Statistical DescriptionPart 3: Similarity, Distance and Data Preparation Before applying data mining techniques, it is important to examine the quality and distribution of a dataset. Real-world datasets often contain missing values, duplicate records, inconsistencies and outliers. These issues can affect statistical ana.. -
[DM Lecture 02] Part 1 Understanding Data: Types and Characteristics Before applying data mining or machine learning techniques, it is essential to understand the structure and characteristics of the data. Different types of data require different methods of analysis, preprocessing and representation. Choosing an appropriate approach begins with identifying what the data represents, how it is measured, and how it is organised. Part1: Understanding Data: Types and..
728x90