2 ms·

What is machine learning?

1 points by mardo5 12y ago

Hello, In the wikipedia entry on Machine learning they mention the 'explicitly programmed instructions' : "Machine learning is a scientific discipline that explores the construction and study of algorithms that can learn from data. Such algorithms operate by building a model based on inputs and using that to make predictions or decisions, rather than following only explicitly programmed instructions." I'm really trying to understand what do these algorithms really are and what do they mean by 'explicitly programmed instructions'. How can you write for example a classifier using the 'explicitly programmed instructions' and how do we compare it to a ML classifier ? Does the ML classifier use heuristics where the 'explicitly programmed instructions' one will look for an optimal solution ? Aren't the 2 just 'explicitly programmed instructions' with different response time ? I read some articles and did some MOOC on ML, what I'm looking for is what do we mean by ML. I'm not that interested in its applications. I'd appreciate any answers, materials, ... Thanks,

stdbrouw 12y ago
Here's a classifier that uses explicitly programmed instructions: def classify(word): return word[0] What you would do instead with supervised learning or other ML methods, is to give the algorithm some inputs and outputs, e.g. "Hello => H", "World => W" and then let the algorithm come up with a model that will know that the answer to "Sunshine" is "S" without you having explicitly coded that. Of course, in reality you wouldn't use machine learning to alphabetize words. It's most useful when explicit or exact classification / modeling / prediction is simply not possible, as e.g. with spam detection.