Application of machine learning in Artificial Intelligence
What is your instant answer if I make a question as “Do you believe machines can think and take its own decision for solving problems as a human?”. The answer should be definitely yes! Why I’m saying like that? Humans learn many new things from childhood step by step. And also in some cases we used our past experiences to map with new experiences. The same scenario applies to machines in machine learning. Simply we can say, machine learning is learning a machine using some training data set and let it to map the previous experiences with the new set of data for decision making purpose. There are several applications of machine learning such as image recognition, speech recognition, Email spam and malware filtering, automatic language translation, product recommendation, virtual personal assistant and etc.
There are different machine learning algorithms varies from their approach, input and output data type and the type of problem that they are going to be solved.
1. Supervised learning model contains the both input and desired output within it. It contains labeled training data. Classification and regression are types of supervised learning.
2. Unsupervised learning model contains only input data set and find the structure or patterns of that input data. When new data comes, it identifies the common features and react based on the present or absent of such common patterns.
Image recognition is used to identify objects, places, persons for various kinds of purposes such as criminal investigation, security monitoring systems and etc. The most common case you experienced with image recognition is automatic friend tagging suggestions in Facebook.
Machines can listen us and understand what we are talking. As well as machines can grab our speech to suggest the related things what we are talking. This is the scenario behind the speech recognition. Few years ago, when we want to search something in Google we wanted to type it and press search button. But now? Just we have to speak. Machines can assist us recognizing our voice. Examples are Siri, Google assistant, Cortana and Alexa.
When we are receiving emails into our inbox, it is filtered as important, normal and spam. What is the technology lies behind this email filtering. Natural language processing techniques are lying behind this scenario.
Nowadays, we don’t need to worry about the languages we are not aware, because automatic translation methods in machine learning can able to convert the different language texts into our familiar language.
Product recommendation is another approach in machine learning used in e-commerce sites such as Amazon, Netflix etc. Machines can identify your interests using some machine learning algorithms and suggest some advertisements according to your interests.
Only referring the above facts we can say how machine learning affects our day-to-day life. There are so many other areas I haven’t mentioned above. So, how can we disagree with the statement above I mentioned? However most machine learning models are data intensive. That mean more and more data get the more and more accuracy for the purpose. Nowadays researches are going through the machine learning models that can achieve the targets using less amount of data.
0 Comments