A good article I saw recently was transferred from the Yunqi community.
The K-means algorithm has a long history and is also one of the most commonly used clustering algorithms. The K-means algorithm is very simple to implement, therefore, it is very suitable for novice machine learning enthusiasts. First, let's review the origin of the K-Means algorithm, and then introduce its more typical application scenarios.
origin
The term "K-means" was first coined in 1967 by James MacQueen in his paper "Some Methods for the Classification and Analysis of Multivariate Observations". In 1957, Bell Labs also used standard algorithms for pulse code modulation techniques. In 1965, EW Forgy published essentially the same algorithm - the Lloyd-Forgy algorithm.
What is the K-Means algorithm?
Clustering is dividing data into groups so that data points in the same group are more similar than data points in other groups. In short, clustering is the division of data points with similar characteristics into groups, that is, into clusters. The goal of the K-means algorithm is to find groups in the data, and the number of groups is represented by the variable K. Each data point is assigned to one of the K groups by an iterative operation based on the characteristics provided by the data. K = 2 in the figure below, so two clusters can be identified from the original dataset.
10 Interesting Use Cases of the K-Means Algorithm
The K-means algorithm is executed on a dataset, and the outputs are:
1. K center points: each center point of the k clusters identified from the dataset.
2. Full labelling of the dataset to ensure that each data point can be assigned to one of the clusters.
Top 10 Use Cases of K-Means Algorithm
The K-means algorithm can usually be applied to data sets with small and continuous dimensions and values, such as grouping the same things from a randomly distributed set of things.
1. Document Classifier
Divide documents into several different categories based on tags, topics, and document content. This is a very standard and classic K-means algorithm classification problem. First, it is necessary to initialize the documents, represent each document as a vector, and use the term frequency to identify common terms for document classification. This step is necessary. The document vectors are then clustered to identify similarities within groups of documents. Here is an example implementation of the K-means algorithm for document classification.
2. Item transfer optimization
Use the combination of K-means algorithm to find the best launch position of UAV and genetic algorithm to solve the problem of traveling salesman's driving route and optimize the UAV item transmission process. Here is the white paper for the project.
3. Identify the crime location
Using crime data related to specific areas in a city, analyzing crime categories, crime locations, and the correlation between the two, allows for high-quality surveys of crime-prone areas in a city or region. This is a paper based on crime data from the Delhi FIR.
4. Customer classification
Clustering helps marketers improve their customer base (working within their target area) and further segment customer categories based on their purchase history, interests or activity monitoring. This is a white paper on how telecom operators divide prepaid customers into recharge models, sending text messages and browsing websites. Categorizing customers helps companies target specific customer groups with specific ads.
5. Team status analysis
Analyzing the state of players has always been a key element in sports. Machine learning is also playing a crucial role in this field as the competition becomes more intense. If you want to create a good team and like to identify similar players based on player status, then the K-means algorithm is a good choice. For details and implementation, please refer to this article.
6. Insurance Fraud Detection
Machine learning also plays a vital role in fraud detection and is widely used in the fields of automotive, health insurance, and insurance fraud detection. Use historical data on past fraudulent claims to identify new claims based on its similarity to the fraudulent pattern cluster. Because insurance fraud can cost companies millions of dollars, fraud detection is critical to companies. This is a white paper on using clustering to detect fraud in auto insurance.
7. Ride data analysis
The public-facing dataset of Uber rides provides us with a wealth of valuable datasets on traffic, transit times, peak ride locations, and more. Analyzing this data will not only be of great benefit to Uber, but it will also help us gain insights into the city's transportation patterns that can help us plan the future of our cities. Here's an article analyzing Uber's data process using a single sample dataset.
8. Network Analysis Criminals
Network analysis is the process of collecting data from individuals and groups to identify important relationships between them. Network analysis is derived from crime files, which provide investigative departments with information to classify criminals at crime scenes. This is a paper on how to cyber-profile network users based on user data preferences in an academic setting.
9. Detailed analysis of call records
Call Detail Records (CDRs) are collections of information about users' calls, text messages and network activity by telecommunications companies. Combining call details with customer profiles can help telcos make more predictions about customer needs. In this post, you'll learn how to use the unsupervised K-Means clustering algorithm to cluster customer activity 24 hours a day to understand customer usage over a period of hours.
10. Automated clustering of IT alerts
Large enterprise IT infrastructure technical components such as network, storage or database generate a large number of alert messages. Since alert messages can point to specific actions, manual filtering of alert messages is necessary to ensure prioritization of subsequent processes. Clustering the data provides insight into alert categories and mean time to repair, which can help predict future failures.
The above is the translation.
This article is translated by Alibaba Cloud Yunqi Community Organization.
The original title of the article "10 Interesting Use Cases For the K-Means Algorithm", translator: Mags, proofreader: Yuan Hu.