Algorithms are everywhere in the world, governing natural processes, human actions and machine’s behavior, and involving rules, laws, regularities, logics, instructions, or computing programs.

Algorithms could be specified as a systematic mathematical procedure/process/technique to produce the solution to a problem/the answer to a question/the prediction to an event/the effect to a cause.

Algorithm: process or set of rules to be followed in action, reasoning, computation, or calculations or other problem-solving operations, especially by a computer.

Basically, the goal of an algorithm is to solve a specific problem, usually defined as a sequence of steps.

Broadly, there are decision procedures, yes/no, and computation procedures.

There are infinite classes of problems, no handwritten algorithms are known, and machines are critical.

One has to make the difference between regular Algorithms, Machine Learning Algorithms and AI Algorithms.

Traditional Algorithm takes some input and applying logic in the form of code, it produces output based on the rules and encoded parameters.

ML algorithm predicts an output on the basis of learning through the input provided to it, called the Training process.

A programmer creating a math model mapping input to output, adjusting its parameters by back-feeding it with input and expected output. Give the algorithm data to learn from and it adjusts parameters to explain the data, and use these set of parameters to explain/predict new data.

A typical mistake is to confuse Machine Learning Algorithms with AI Algorithms.

To be applied to clustering, classification and regression problems, AI algorithms enable the software to model the world in terms of causal relationships between independent and dependent entity variables, categorical and continuous, quantitative and qualitative.

Machine Learning is made up of a series of algorithms, like this list of 10 common Machine Learning Algorithms:

Linear Regression. ...

Logistic Regression. ...

Decision Tree. ...

SVM (3.Support Vector Machine) ...

Naive Bayes. ...

KNN (K- Nearest Neighbors) ...

K-Means. ...

Random Forest,

Dimensionality Reduction Algorithms,

Gradient Boosting & AdaBoost

It combines multiple weak or average predictors to build a strong predictor, and used in data science competitions like Kaggle, AV Hackathon, CrowdAnalytix.

These are the most preferred machine learning algorithms today, used along with Python and R Codes to achieve accurate outcomes.

The 10 Algorithms: Machine Learning Engineers Need to Know

Machine Learning is designed to learn in the same way as a child, by trial and error empirical ways, with feedback loops and reward-punishment learning. Having a large dataset, a ML can find data patterns and builds hypothesis/theories/assumptions based on those findings.

Machine learning: set of algorithms that enable the software to update and “learn” from previous outcomes without the need for programmer intervention. It is fed with structured data in order to complete a task without being programmed how to do so.

Being data-intensive, machine Learning algorithms can help computers recognize objects, play chess, perform surgeries, or get smarter and more personal, while failing to any self-identity or self-awareness.

What is an AI Algorithm?

Briefly, all machine learning algorithms are badly biased, not mentioning how training, testing and validating data sets artificially created.

If you are after real AI systems, just don’t use “TOP 10 Machine Learning Algorithms” as sampled below with some comments.

TOP 10 Machine Learning Algorithms – Good Audience

I can’t help but wonder about the degree to which the biased machine learning algorithms employed by Facebook, Google, Tinder, Amazon, IBM, Microsoft, Chinese BAT, etc. fooling their billions users, as noted here:

Kiryl Persianov's answer to Is machine learning biased because people are biased ?

https://www.quora.com/What-is-an-algorithm-in-AI/answer/Kiryl-Persianov