Chapter 1. Meet Apache Mahout
Figure 1.1. Apache Mahout and its related projects within the Apache Software Foundation
Figure 1.2. A recommendation from Amazon. Based on past purchase history and other activity of customers like the user, Amazon considers this to be something the user is interested in. It can even list similar items that the user has bought or liked that in part caused the recommendation.
Figure 1.3. A sample news grouping from Google News. A detailed snippet from one representative story is displayed, and links to a few other similar stories within the cluster for this topic are shown. Links to all the stories that are clustered together in this topic are available too.
Figure 1.4. Spam messages as detected by Yahoo! Mail. Based on reports of email spam from users, plus other analysis, the system has learned certain attributes that usually identify spam. For example, messages mentioning “Viagra” are frequently spam—as are those with clever misspellings like “v1agra.” The presence of such terms is an example of an attribute that a spam classifier can learn.
Chapter 2. Introducing recommenders
Figure 2.1. Relationships between users 1 to 5 and items 101 to 107. Dashed lines represent associations that seem negative—the user doesn’t seem to like the item much but expresses a relationship to the item.
Figure 2.2. Simplified illustration of component interaction in a Mahout user-based recommender