AI
-
Assumptions of K-Means Clustering – Part 1
K-Means Clustering is a popular unsupervised machine learning algorithm that groups data into a predefined number of clusters based on similarity. However, like every algorithm, K-Means makes certain assumptions about the data and the structure of the problem. Understanding these assumptions is crucial to applying…
-
Beyond K-Means: Other Notions of Distance
K-Means clustering is a great starting point for understanding unsupervised machine learning. However, it comes with important limitations, especially when data is complex or doesn’t conform to its assumptions. In this article, we explore what lies beyond K-Means — alternative methods and concepts that better…
-
Examples of K-Means Clustering Problems
K-Means clustering is one of the simplest and most widely used unsupervised machine learning algorithms. Its goal is to group data points into clusters based on similarity. In this article, we’ll explore different types of K-Means clustering problems, challenges related to choosing the right number…
-
Making Sense of Unstructured Data: How to Evaluate Clustering?
Clustering is one of the most popular techniques for exploring and organizing unstructured data—like text, images, or customer behavior. But once you apply a clustering algorithm, how do you know if it actually worked well? In this post, we’ll break down the basics of clustering…
-
Making Sense of Unstructured Data: Graph Theory
Unstructured data is often described as any data that does not have a predefined structure or is not easily organized into rows and columns like structured data. Examples include text, images, audio, and social media posts. This kind of data is abundant in today’s digital…
-
Making Sense of Unstructured Data: K-means Clustering
Data is everywhere. From social media posts to customer reviews, much of the data we generate and collect is unstructured. Unlike structured data (like spreadsheets), unstructured data doesn’t follow a predefined format, making it harder to analyze. That’s where machine learning techniques like K-means clustering…
-
Making Sense of Unstructured Data: Distance and Scaling Measures
In the age of big data, unstructured data like text, images, audio, and videos make up the bulk of the information we generate. Unlike structured data (like spreadsheets), unstructured data doesn’t come in neat rows and columns. To analyze it effectively, especially for machine learning…
-
Making Sense of Unstructure Data: Dimensionality Reduction (PCA & tSNE)
What is Dimensionality? In data science, dimensionality refers to the number of features (also called variables or attributes) in a dataset. For example, an image that’s 28×28 pixels has 784 dimensions (28 multiplied by 28), because each pixel represents a feature. High-dimensional data presents challenges:…
-
Making Sense of Unstructured Data: Unsupervised Learning
In today’s data-rich environment, most of the information we encounter is unstructured. From social media posts and support tickets to satellite imagery and audio recordings, unstructured data surrounds us. Making sense of this kind of data is critical for businesses, scientists, and engineers alike. One…
-
Making Sense of Unstructured Data: Kullback-Leibler (KL) Divergence
In today’s data-driven world, most of the information we encounter is unstructured. From social media posts and emails to images and audio recordings, unstructured data surrounds us. Understanding and making sense of this data is crucial for businesses and researchers alike. One fundamental concept that…