Extracting relationships

There are a number of techniques available to extract relationships. These can be grouped as follows:

• Hand-built patterns
• Supervised methods
• Semi-supervised or unsupervised methods
• Bootstrapping methods
• Distant supervision methods
• Unsupervised methods

Hand-built models are used when we have no training data. This can occur with new business domains or entirely new types of projects. These often require the use of rules. A rule might be:

"If the word "actor" or "actress" is used and not the word "movie" or "commercial", then the text should be classified as a play."

However, this approach takes a lot of effort and needs to be adjusted for the actual text in-hand.

If only a little training data is amiable, then the Naive Bayes classifier is a good choice. When more data is available, then techniques such as support vector machine (SVM), regularized logistic regression, and random forest can be used.

Although it is useful to understand these techniques in more detail, we will...