Skip to main content

Introduction to Machine Learning – ML

Machine Learning – ML

ML is one of the hot buzz word now, let me explain what it is and why it is? In traditional programming, we feed data and program to a computer to get output, but in Machine learning, we feed data and output to a system to generate algorithm or programming for that data to find a pattern. So that we can forecast the results of new set data by applying the pattern which has found on existing data. This is a simple definition to understand ML concept. ML gives computer systems the ability to learn the data, without being programmed explicitly. Day to day growing internet traffic makes human thinking and analysis of data to limited. To do this we need a machine which can analyze the huge data and find a pattern on that data. This is what machine learning does. It is a continuously developing field. Example: Facial recognition technology allows social media platforms to help users tag and share photos of friends. Python if a popular language for ML followed by R programming language. As a field, machine learning is closely related to computational statistics, so having a background knowledge in statistics is useful for understanding and leveraging machine learning algorithms. Concepts like correlation and regression, are commonly used techniques for investigating the relationship among quantitative variables. Most important thing is to which kind of industry we are applying ML and ML programmer/ engineer must get the proper domain knowledge to apply ML on that system or application engineer or core domain people must support to analyze the pattern on the data.  There are different approaches to ML such as k-nearest neighbor, decision tree learning etc.

Machine learning methods:

In machine learning, tasks are generally classified into broad categories. These categories are based on how learning is received or how feedback on the learning is given to the system developed.
Two of the most widely adopted machine learning methods are supervised learning which trains algorithms based on example input and output data that is labeled by humans, and unsupervised learning which provides the algorithm with no labeled data to allow it to find structure within its input data.  As of now, these definitions might look complicated. In later articles I will talk more about on these two methods.

Because machine learning is a field that is continuously being innovated, it is important to keep in mind that algorithms, methods, and approaches will continue to change.




Comments

Post a Comment

Popular posts from this blog

Introduction to latest technologies such as IoT, IoE, IIoT,WoT

Hi Everyone, Now-a-days, we are hearing terms such as IoT, IoE, IIoT, WoT. In this article i am trying to explain what exactly these terms means.  Let me start with fancy word called IoT (Internet Of Things). Its a world of things (such as vehicles, home appliances, any physical device) connected.  These devices embedded with firmware, software, electronics, sensors, actuators and connectivity which makes any physical device intelligent, so that they can talk with each other. Do you want to know why? I will be explaining more details in my next article. IIoT is Industrial IoT, i.e using IoT in manufacturing industries. This brings brilliant machines together, advanced analytics and many more. It has great potential to change the way how the Industry works and it enables M2M machine-to-machine communication. Next article i will tell some real time applications of this. IoE is Internet of Everything where people, process, data and things(these are ...
2 comments
Read more

Introduction to Deep Learning

Deep Learning is a type of machine learning in which a model learns to perform classification tasks directly from images, text or sound. Its usually implemented using a neural network architecture. Here deep refers to the "Number of layers in the network" i.e the more layers, the deeper the network. Traditional neural network contain only 2-3 layers, but in deep network contains hundreds of layers. Deep learning example is  A self-driving vehicle slows down as it approaches a pedestrian network . Deep learning is especially well-suited to identification applications such as face recognition, text translation, voice recognition and advanced driver assistance systems and traffic sign recognition etc. Deep learning is growing faster because it really works.  What is the Difference Between Deep Learning  and Machine Learning? Deep learning is a sub-type of machine learning. For example with machine  learning, you manually extract the relevant features of...
Post a Comment
Read more

Introduction to IoT ecosystem

Basics of IoT IoT (Internet of Things) refers to a network where devices communicate to devices and devices communicate to software applications over the internet . This web-enabled devices/internet enabled devices will usually collect data, send and act upon data from their surrounding environment. •       The IoT network consists of Edge side (sensors, actuators and controllers connected to the internet) and Cloud side. •       The devices on the Edge side can either connect to the internet directly or using an Edge (IoT) gateway . The Edge gateway is used for: •        Providing a connection to the internet for devices that cannot connect to the internet on their own. These devices don’t have feature to connect to the Internet. •        Local functionality/control that must work even without internet availability •       ...
Post a Comment
Read more