Don’t know whether god would help or not ,but machine can !!
Before Learning of machine ,We must learn how machine learn its learnings
A computer program is said to learn from experience ‘E’ with respect to some task ‘T’ and some performance measure ‘P’, if its performance on ‘T’ ,as measured by ‘P’ improves with experience E.
Let’s make it more simple for this child
Machine Learning is the science of getting computers to learn without being explicitly programmed . At its core, is a set of statistical methods meant to find patterns of predictability in data sets.
Let’s Start by distinguishing between automation problems and learning problems. Machine learning can help automate your processes, but not all automation problems require learning.Automation without learning is appropriate when the problem is relatively straightforward.
If you are a product company ,you want machine learning to predecit some of your most vital questions like i)Do we need to increase our stock in next month or not ii)Which should be our targeted users so on and so fourth.
If you are a consultancy then your target is to be answer this question as appropritae as you can.
Before investment or implementing machine learnings we should be aware of the fact ,when to use machine when to not .Like there is no need to use machine learning for computing a payroll : for this task we just need an algorithm.
Let’s jump to the next chapter :
Data/Big Data : Machine Learnings is just a normal machine till it starts learnings and to learn it needs data , a lots of data by which it can learn by itself and then answer your query based on the learning on your data set.
So can i say if i provide data to machine ,it can answer my questions ?
Answer is both yes or no .Let’s clarify it .
(But don’t ask this type of questions 🙂
Let’s say you have user data like
customerName: jon snow , Gender :male ,Address: Game of thrones , Designations : King.
By providing this type of data ,if you want to predict whether your customer will buy your product in next month or not then you are on the wrong track and considering machine learning as the magic. It means you should provide relevant data that have relationship with the outcome.
Let’s come to the real problems :
If your data set contains some informations by which algorithm can trace or find some patterns of the behaviour of some occurence then it will always answer some problem that includes :Detecting Spam,Product recommendation,Medical Diagnosis,Customer segmentation and Lifetime value prediction,Financial analysis
,Predictive maintenance,Image recognition. In other way it will answer the questions based on previous experiances and always learn with the help of data
, more the data you provide , more the accurate it will become .