Machine Learning

Supervised or unsupervised learning? Given a dataset of patients diagnosed as either having diabetes or not, learn to classify new patients as having diabetes or not.

supervised

If you have software to examine individual accounts and for each account you want to decide if it's been hacked, is that a regression or classification problem?

Classification

What is Machine Learning?

Teaching computers to solve problems on their own.

What is unsupervised learning?

We are given data without labels. The program has to find structure within the data.

What is supervised learning? What are the two categories of supervised learning problems?

We give the algorithm a dataset in which "right answers" are given, that the program uses to predict future answers. For example, giving it a graph with specified values, and asking it to predict the values of other, non specified values. The two categories are regression and classification

Supervised or unsupervised learning? Given a database of customer data, automatically discover market segments and group customers into different market segments

unsupervised

Supervised or unsupervised learning? Given a set of news articles found on the web, group them into set of articles about the same story

unsupervised

What is Octave?

A prototyping tool for learning algorithms that allow faster creation than Java or C++

Suppose your email program watches emails you do or do not mark as spam, based on that it better learns to spam. If Machine learning is defined as "learn experience E with some respect to task T and some performance measure P, if it's performance on T as measured by P improves with experience E" What is T?

Classifying emails as spam or not spam. The number it correctly classifies is P. And the experience is watching you label email as spam or not spam.

What is the difference between Machine Learning and AI?

Machine learning grew out of AI, so it's a subset of AI.

What is a regression problem?

Predicting a value on a continuos value chart.

If you have a large inventory of identical products and you want to predict how many of these items will sell over the next three months, is it a classification or regression problem?

Regression

Supervised or unsupervised learning? Given email labeled spam/not spam, learn a spam filter.

Supervised

What is a classification problem?

There are two (or more) output choices, and you're trying to figure out which output will occur. It's best to define the output choice not as an axis, but a symbol.