Data Science - Software Developer's Tour

Understanding K-Nearest-Neighbors (KNN) – Essential Machine Learning Algorithm

Paweł Majewski — Sun, 29 Jun 2025 01:21:34 +0000

A quick introduction to this algorithm

KNN is used to determine the class of our sample data. To understand this more clearly, please take a look at the plot below. We have two groups of samples: blue ones and red ones. We know a priori which samples are blue and which are red. Now, however, a new sample has arrived – a green one. Our task is to determine whether this sample better fits with the reds or the blues.

As you might guess, the simplest way to assign class A or B to a new sample is to calculate the distance between our sample and other points. Why does KNN have a ‘K‘ in its name? The ‘K‘ represents the number of nearest neighbors used to predict a classification.

How do we calculate distance between samples?

The easiest way to calculate the distance is to use a formula we all know from primary school, called the Euclidean distance. Of course, there are other algorithms that are better depending on the distribution of our data.

\sqrt{{(x ₂ - x ₁)}^{2} + {(y ₂ - y ₁)}^{2}}

Just code this function

def euclidean_distance(a, b):
  return np.sqrt(np.sum(a - b) ** 2)

Now, we need to find the K nearest neighbors. First, we need to calculate the distance between each point and the test point.

def get_neighbors(X, y, test_point, k):
    distances = []
    for i in range(len(X)):
        distance = euclidean_distance(X[i], test_point)
        distances.append((X[i], y[i], distance))
    distances.sort(key=lambda x: x[2])
    neighbors = distances[:k]
    return neighbors

In this step, we will calculate class for our test point.

This is the result of get_neighbors function.

As you can see, we have three instances of class 0 and two instances of class 1. Therefore, when the KNN algorithm is run on this data with K equal to 5, it will classify the test point as class 0, since it is nearest to this class.

This is how it looks like

Our test point has three connections with the blue ones and only two connections with the red ones.

d(P, Q) = \sqrt{\sum i_{1}^n ((p_i - q_i)^2)}

Our function for calculating the Euclidean distance, called euclidean_distance, is designed to solve real-life problems that are not limited to two dimensions but can also handle n-dimensional scenarios.

The post Understanding K-Nearest-Neighbors (KNN) – Essential Machine Learning Algorithm first appeared on Software Developer's Tour.

A Step-by-Step Guide to Web Scraping for Beginners

Paweł Majewski — Sat, 21 Jun 2025 08:38:00 +0000

In this article, I would like to raise the subject of web scraping.

Before we start, we have to consider web scraping on several layers: legal and ethical.

Let’s tackle the topic of the legality of web scraping first.

Fundamentally, the law depends on the country you live in. You should remember about:

Authors’ right

Terms of Service

General Data Protection Regulation (GDPR, European Union)

Computer Fraud and Abuse Act (CFAA, USA)

Outside the regulations you have to make sure that you are not affecting on target’s infrastructure by for example DDOSing target’s site.

Let’s get to practice

We want to get all titles from the blog section.

When we analyze the blog section, we can see that the title is an a tag inside a div with the class blog-page-title.

The steps we need to take to accomplish our task:

1. Go to the website.
2. Wait for the page to load until there is a div with the class “blog-page-title” and an a tag inside it.
3. Find the a tag inside the div with the class mentioned above.
4. Print the text of these a tags.

from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC

url = "https://pawelmajewski.com/blog"
driver = webdriver.Chrome()
driver.get(url)
div_class = "blog-page-title"
wait = WebDriverWait(driver, 3)
wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, f"div.{div_class} a")))
a_tags = driver.find_elements(By.CSS_SELECTOR, f"div.{div_class} a")

for a_tag in a_tags:
    print(a_tag.text)

driver.quit()

I recommend using the Selenium library both for writing your own web scrapers as well as for writing, for example, integration tests.

The post A Step-by-Step Guide to Web Scraping for Beginners first appeared on Software Developer's Tour.