Machine Learning

In the era of AI development and algorithm, proper tool can change your business for ever.

ML is a Scientific study of algorithms and statistical models that computer systems use to effectively perform a specific task without using explicit instructions, relying on patterns and inference instead.

It is seen as a subset of artificial intelligence. Machine learning algorithms build a mathematical model of sample data, known as “training data”, in order to make predictions or decisions without being explicitly programmed to perform the task.

Machine learning algorithms are used in a wide variety of applications, such as email filtering, detection of network intruders, and computer vision, where it is infeasible to develop an algorithm of specific instructions for performing the task. Machine learning is closely related to computational statistics, which focuses on making predictions using computers. The study of mathematical optimization delivers methods, theory and application domains to the field of machine learning. Data mining is a field of study within machine learning, and focuses on exploratory data analysis through unsupervised learning. In its application across business problems, machine learning is also referred to as predictive analytics.  for more see – Wikipedia

Comparison:

To better understand the differences, interactions and similarities with neural network here is a little side to side info about both of them to clear things up.

Neural Network

Required skills:

Probability, Programming Skills, Data structures and Algorithms, Knowledge about machine learning frameworks, Big Data

Applied in:

  • Health care
  • Retail
  • E-commerce
  • Online recommendations
  • Tracking price changes
  • Better costumer service and delivery

Siri, Google Maps, Search engines etc.

Neural Network

Required skills:

Probability and statistics, Data modeling, Programming skills, Data structures and algorithms, Mathematics, Linear algebra and graph theory

Applied in:

  • Finance
  • Health Care
  • Retailing
  • Machine learning
  • Artificial Intelligence
  • Stock Exchange Prediction

Image Recognition, Image Compression and search, Search engines, self driving etc.

The Process of machine learning and algorithms

Step 1

Gathering date from various sources.

Step 2

Cleaning data to have homogeneity.

Step 3

Model Building – Selecting the right ML algorithm

Step 4

Gaining Insights from the model’s results

Step 5

Data Visualization – Transforming results into visual graphs.

Linear Regression

111-01

Description

Advantages

Disadvantages
The”Best fit” line through all data points. Predictions are numerical. Easy to understand – you clearly see what the biggest drivers of the model are.
  • Sometimes too simple to capture complex relationships between variables.
  • Doeas poorly with correlated features

Logistic Regression

111-05
The adaptation of linear regression to problems of classification ( e.g. m yes/no questions, groups etc. Easy to understand.
  • Sometimes too simple to capture complex relationships between variables.
  • Doeas poorly with correlated features

Decision Tree

111-03
A series of yes/no rules based on the features, forming a tree, to match all possible outcomes of a decision. Easy to understand.
  • Not often used on its own for predictions because it’s also often too simple and not powerful enough for complex data

Random Forest

111-02
Takes advantages of many decision tree, with rules created from sub-samples of features, each tree is weaker than a full decision tree, but by combining them we get better overall performance. a sort of “wisdom of the crowd”. Tends to result in very high quality models. Fast to train
  • Models can get very large
  • Not easy to understand predictions

Gradient Boosting

111-04
Uses even weker decision trees, that are increasingly focused on”Hard” examples. High – Performing.
  • A small change in the feature set or training set can create radical changes in the model.
  • Not easy to understand predictions.

Neural Network

111-06
Interconnected <neurons: that pas messages to each other Deep learning uses several layers of neural networks stacked on top of each other. Can handle extremely complex tasks- no other algorithm comes close in image recognision.
  • Bery slow to train, because they often have a very complex architecture
  • Almost impossible to understand predictions

Read more:

Here are my other articles you might be interested in