If you’re into a particular technology you’re likely to read, vote or download content surrounding that technology. By looking at patterns of behviour you can make an educated guess about what one person may like based on how similar or disimilar their actions are to others.
We’ve generated some user data that revolves around a set of articles. Each article has one or more tags, and over the course of 30 days users are reading, upvoting or downvoting articles as well as downloading files assoicated with the articles.
Your job is to decide how important each action is. Is an upvote more significant than a file download? Is a downvote important? What about the tags associated with each article? The viewing habits? How does it all fit together?
Build a model that can input this data and make a decision on whether a given user would .be interested in a given article, and that, given a user, can suggest articles most likely to be of interest.
- Java 8
- Maven