Table of contents

Key takeawaysCollaboration platforms are essential to the new way of workingEmployees prefer engati over emailEmployees play a growing part in software purchasing decisionsThe future of work is collaborativeMethodologyMachine learning algorithms are the engines of machine learning, meaning it is the algorithms that turn a data set into a model. Which kind of algorithm works best (supervised, unsupervised, classification, regression, etc.) depends on the kind of problem you’re solving, the computing resources available, and the nature of the data.

An algorithm in machine learning is a procedure that is run on data to create a machine learning model.

Machine learning algorithms perform pattern recognition. Algorithms learn from data, or are fit on a dataset.

For example, we have algorithms for classification, such as k-nearest neighbors. We have algorithms for regression, such as linear regression, and we have algorithms for clustering, such as k-means.

You can think of a machine learning algorithm like any other algorithm in computer science.

For example, some other types of algorithms you might be familiar with include bubble sort for sorting data and best-first for searching.

As such, machine learning algorithms have a number of properties:

- Machine learning algorithms can be described using math and pseudocode.
- The efficiency of machine learning algorithms can be analyzed and described.
- Machine learning algorithms can be implemented with any one of a range of modern programming languages.

You are also likely to see multiple machine learning algorithms implemented together and provided in a library with a standard application programming interface (API). A popular example is the scikit-learn library that provides implementations of many classification, regression, and clustering machine learning algorithms in Python.

Linear regression is a supervised learning algorithm and tries to model the relationship between a continuous target variable and one or more independent variables by fitting a linear equation to the data.

Support Vector Machine (SVM) is a supervised learning algorithm and mostly used for classification tasks but it is also suitable for regression tasks.

The data points are not always linearly separable like in the figure above. In these cases, SVM uses kernel trick which measures the similarity (or closeness) of data points in a higher dimensional space in order to make them linearly separable.

Naive Bayes is a supervised learning algorithm used for classification tasks. Hence, it is also called Naive Bayes Classifier.

Naive bayes assumes that features are independent of each other and there is no correlation between features. However, this is not the case in real life. This naive assumption of features being uncorrelated is the reason why this algorithm is called “naive”.

Logistic regression is a supervised learning algorithm which is mostly used for binary classification problems. Although “regression” contradicts with “classification”, the focus here is on the word “logistic” referring to logistic function which does the classification task in this algorithm. Logistic regression is a simple yet very effective classification algorithm so it is commonly used for many binary classification tasks. Customer churn, spam email, website or ad click predictions are some examples of the areas where logistic regression offers a powerful solution.

K-nearest neighbors (kNN) is a supervised learning algorithm that can be used to solve both classification and regression tasks. The main idea behind kNN is that the value or class of a data point is determined by the data points around it.

A decision tree builds upon iteratively asking questions to partition data. The aim of the decision tree algorithm is to increase the predictiveness as much as possible at each partitioning so that the model keeps gaining information about the dataset. Randomly splitting the features does not usually give us valuable insight into the dataset. Splits that increase purity of nodes are more informative. The purity of a node is inversely proportional to the distribution of different classes in that node. The questions to ask are chosen in a way that increases purity or decrease impurity.

Random forest is an ensemble of many decision trees. Random forests are built using a method called bagging in which decision trees are used as parallel estimators. If used for a classification problem, the result is based on majority vote of the results received from each decision tree. For regression, the prediction of a leaf node is the mean value of the target values in that leaf. Random forest regression takes mean value of the results from decision trees.

GBDT is an ensemble algorithm which uses boosting method to combine individual decision trees.

Boosting means combining a learning algorithm in series to achieve a strong learner from many sequentially connected weak learners. In case of GBDT, the weak learners are decision trees.

Clustering is a way to group a set of data points in a way that similar data points are grouped together. Therefore, clustering algorithms look for similarities or dissimilarities among data points. Clustering is an unsupervised learning method so there is no label associated with data points. Clustering algorithms try to find the underlying structure of the data.

Partition-based and hierarchical clustering techniques are highly efficient with normal shaped clusters. However, when it comes to arbitrary shaped clusters or detecting outliers, density-based techniques are more efficient.

DBSCAN stands for density-based spatial clustering of applications with noise. It is able to find arbitrary shaped clusters and clusters with noise (i.e. outliers).

PCA is a dimensionality reduction algorithm which basically derives new features from the existing ones with keeping as much information as possible. PCA is an unsupervised learning algorithm but it is also widely used as a preprocessing step for supervised learning algorithms.

There is a wide variety of machine learning algorithms that can be grouped in three main categories:

Supervised learning algorithms model the relationship between features (independent variables) and a label (target) given a set of observations. Then the model is used to predict the label of new observations using the features. Depending on the characteristics of target variable, it can be a classification (discrete target variable) or a regression (continuous target variable) task.

In this algorithm, we do not have any target or outcome variable to predict / estimate. It is used for clustering population in different groups, which is widely used for segmenting customers in different groups for specific intervention. Examples of Unsupervised Learning: Apriori algorithm, K-means.

Reinforcement learning works based on an action-reward principle. An agent learns to reach a goal by iteratively calculating the reward of its actions. Using this algorithm, the machine is trained to make specific decisions. It works this way: the machine is exposed to an environment where it trains itself continually using trial and error. This machine learns from past experience and tries to capture the best possible knowledge to make accurate business decisions.

Share