LO 7.1.4.C

Learning Objective: Describe topic models.

Review:

• Topic models.

To facilitate the analysis of our learning models we need to devise a topic layer above the “bags of words” we have created. Each document constitutes a sequence of words, but instead of the words being used directly by the final classifier, the words map to one or more topics (topic model). The topics also are learned from the data (often via unsupervised data mining).

One advantage of using topics is that they make a search engine more efficient. A query can use terms that do not exactly match the specific words of a document. However, if they map to the correct topic/s, the document will still be considered relevant to the search.

An example of a topic layer is shown in the figure below.

Source: Assigned reading