what is Machine Learning ?
Machine learning is a method of data analysis that uses prediction algorithm to get the unknown details from known data’s .
Another way to think about machine learning is used to find pattern in existing data , then create the algorithm and used that model that recognizes those pattern in new data..
What Machine Learning does ?
Finds the pattern in data.
Recognizes those pattern when you identify from data
|| Amount Utilized
|| Issued Location
|| Where it’s Used
So, for example, suppose I have data about credit card transactions. Suppose I have only Five records, each one has four fields and based on this data we have to figure out fraudulent transactions.
Now we have to find out the pattern to check the fault transaction . In the above data’s , three transaction are identified as fault transaction. In this case if you lookout our data’s three transaction ( Vasanth , Dinesh and Devi ). card was issued in location india and it was used from different location .
But once again, do we know that that pattern is truly predictive? Probably not. To check the pattern is predictive We don’t have enough data. To do this well, you’d have enough data that people just can’t find the patterns.
That’s where machine learning comes in.
Why Machine Learning is so important now ?
Why Machine Learning is so important now – Currently we are lived in BigData century and we are having more and more data’s which produced by this world on daily basis . Now question will be raised here , how we are going to unitized those data’s. Applications can rely on models created via machine learning to make better predictions for data’s .If you can solve important business problems with machine learning, you can save a lot of money. There’s real business value there, and so good data scientists who know all three of these things,
statistics, machine learning, and a problem domain.
Different Types of Machine Learning ?
All ML tasks can be classified in several categories, the main ones are:
Supervised Machine Learning —-> The value you want to predict is in the training data.The data is labeled;
Unsupervised Machine Learning —>The value you want to predict is not in the training data.The data is unlabeled ;
When to Use Machine Learning ??
- Difficult for humans to express rules.
- A large amount of historical data is available.
- Patterns/Relationships are dynamic .