DEV Community

Cover image for What is Machine Learning?
Burak Boduroğlu
Burak Boduroğlu

Posted on

What is Machine Learning?

Machine learning is a subset of artificial intelligence (AI) that focuses on the development of algorithms and statistical models that enable computers to improve their performance on a specific task through experience. Instead of being explicitly programmed to perform a task, a machine learning system learns from data and adapts its behavior over time.

General Process of Machine Learning

1. Data Collection: Collecting the right information for the machine learning model to learn is really important. The amount and quality of the data are key factors in how well the model will work.

2. Data Preprocessing: Getting the data ready for analysis is crucial. This means fixing missing values, making sure the data is in a standard form, and doing other tasks to make sure it works well with the chosen algorithm.

3. Feature Extraction: Choosing the right features or variables is important. This step helps simplify things and ensures the model pays attention to the most important aspects of the data, reducing complexity.

4. Model Training: At the heart of machine learning is training the model with a dataset to teach it how to make predictions or decisions. While training, the model fine-tunes its internal settings to minimize the gap between its predicted results and the actual outcomes in the training data.

5. Model Evaluation: After training, we evaluate the model's performance using a different dataset that it hasn't seen before. We use various metrics like accuracy, precision, recall, and F1 score, depending on the type of problem (classification, regression, etc.), to measure how well the model is doing.

6. Model Tuning: Fine-tuning the model involves adjusting its hyperparameters or structure to enhance performance. This could mean tweaking parameters, experimenting with different algorithms, or testing various feature sets to find the optimal configuration.

7. Prediction/Inference: After the model is trained and assessed, it's ready to be employed for making predictions or decisions on fresh, unseen data.

Three Categories of Machine Learning Algorithms

1. Supervised Learning: The model learns from a labeled dataset, where input data is matched with corresponding output labels. The objective is to grasp the relationship between inputs and outputs, enabling the model to make predictions on new, unseen data.

2. Unsupervised Learning: In unsupervised learning, the model receives input data without explicit output labels. Its goal is to uncover patterns, relationships, or structures within the data. This may involve tasks like clustering similar data points or reducing dimensionality.

3.Semi-Supervised Learning: This type of ML is a combination of supervised and unsupervised learning. The algorithm is trained on a mix of labeled and unlabeled data, allowing it to learn from both types of data.

4. Reinforcement Learning: In reinforcement learning, the model learns through interaction with an environment and receives feedback in the form of rewards or penalties. The objective is for the model to learn a policy that maximizes cumulative rewards over time.


Machine learning finds applications in diverse fields such as image and speech recognition, natural language processing, recommendation systems, and autonomous vehicles, among others. The success of a machine learning model relies on the quality of data, the selection of the algorithm, and the fine-tuning of parameters.


Top comments (6)

proteusiq profile image
Prayson Wilfred Daniel • Edited

🤗 It's exciting to observe your interest in the field of Machine Learning. I've noticed a few inaccuracies in the diagram you've shared.

Incorrect Diagram

Firstly, it's important to clarify the distinction between Logistic Regression and Linear Regression. Logistic Regression is primarily used for classification tasks, where the objective is to categorize data into distinct classes. On the other hand, Linear Regression is typically employed for regression tasks, aiming to predict a continuous output based on input variables.

Regarding KNN (K-Nearest Neighbors), Tree-based models (like Decision Trees), and Neural Networks, it's worth noting that these methods are versatile and can be applied to both classification and regression problems. This flexibility makes them valuable tools in a variety of Machine Learning tasks.

Furthermore, it seems there's a bit of confusion about the 'basic' and ʼother variationsʼ category in your diagram. Variations such as online learning and transfer learning are not distinct model types per se. Rather, online learning refers to a method of training models incrementally by continuously feeding data, which could include any of the models mentioned. Transfer learning, in contrast, involves taking a neural network that has been trained on one task and repurposing it for a different but related task, by transferring the learned features (weights).

Deep Learning, another critical aspect of your diagram, is essentially an advanced form of Neural Networks, typically characterized by having more than three layers. These multiple layers enable the model to learn complex patterns and perform high-level feature extraction, making Deep Learning a powerful tool in many applications.

In the realm of Reinforcement Learning, Deep Learning has become a dominant methodology. This approach combines the principles of Deep Learning with the decision-making capabilities of Reinforcement Learning, leading to sophisticated models capable of learning from their environment and making optimal decisions over time.

I hope this explanation helps to clarify the distinctions and correct the inaccuracies in your diagram. I'm genuinely enthusiastic about your journey in Machine Learning and am rooting for your continued growth and success in this exciting field. 🌟 Keep up the great work!

burakboduroglu profile image
Burak Boduroğlu

Thank you for this clarifying comment I will be removing the diagram from my post. Thank you very much for contributing and helping with my post.🤗

proteusiq profile image
Prayson Wilfred Daniel • Edited

Always welcome 🙃. Keep on building your understanding of ML. It is such a beautiful and rewarding career path.

BTW: You can use perplexity ai to assist you: Example: We could ask one by one the following question:

  1. What is Machine Learning?
  2. What are the steps in doing ML?
  3. What are the types of ML?
  4. List in a table like algorithms in their respective ML Types.

With the information and resources, we could read more, try to play with ML, perhaps with simple drag and drop UI such as orange 🍊

Thread Thread
burakboduroglu profile image
Burak Boduroğlu • Edited

Thanks again.🙏🏼🤗

What is differences between "ChatGPT" and "perplexity"? Is "perplexity" more powerful AI tool?

Thread Thread
proteusiq profile image
Prayson Wilfred Daniel • Edited

ChatGPT, with its GPT-4 models, offers the most advanced and most power Genarative AI. It is currently at the throne of LLMs.

Perplexity has developed their own model, pplx, which effectively gathers online data to generate responses, akin to RAG(Retrieval-Augmented Generation). RAG differentiates itself by citing sources for the information it provides, offering users the opportunity for further exploration, a feature not currently available in ChatGPT.

Thread Thread
burakboduroglu profile image
Burak Boduroğlu

I understood. This is exciting! 😊

Some comments have been hidden by the post's author - find out more