chaps2's sandbox: Browsing History Flag Recommender

Browsing History Flag Recommender is a feature that adds content recommendations based on users' flag-based browsing history. It is written as an add-on feature to the Browsing History feature and is designed to provide example use of the Flag Recommender module. It provides the following:

• Adds a "Users who viewed this item also viewed" view block to node pages included in the "Browsing History Content" context. All users see the same recommendations.
• Adds a "You may also like" view block to the "My browsing history" page. Each user sees different recommendations.
• Works for both authenticated and anonymous users.
• Views can be customised and themed to suit.

Dependencies
This module depends on the Browsing History feature and the Flag Recommender module (and hence Recommender API).

Configuration
For the views to work, the patch in #673786: add views support must be applied to the Recommender API module.
Each view has a recommender app id filter that may need to be changed as follows:

1. Navigate to the flag recommender configuration page at Administer > Site Configuration > Flag recommender.
2. If the application id for "Browsing history content itemt2item" is not "1", or if the Application ID for 'Browsing history content anonymous item 2 item" is not "2" then make a note of them, otherwise you don't need to do anything.
3. Edit the view named "content_predictions2" and change the default "Recommender: Application ID" filter value to match the Application ID for "Browsing history content itemt2item". Do the same for the "Block anon" display of the same view but set the filter value to match the Application ID for "Browsing history content anonymous item 2 item".
4. Edit the view named "content_similarity" and change the "Recommender: Application ID" filter value to match the Application ID for ""Browsing history content itemt2item".

The "Users who viewed this item also viewed" block is based on recommendations determined from authenticated users' browsing history. If your site has mainly anonymous users you may get better recommendations based on anonymous user browsing history. To do this, edit the "content_similarity" view and change the value of the "Recommender: Application ID" filter to match the Application ID listed for the "Browsing history content anonymous item 2 item" flag recommender configuration.