1
Introducing Oracle9iAS Personalization

This chapter introduces Oracle9iAS Personalization, describes what it does and how it does it, and describes its main components.

What Is Oracle9iAS Personalization?

Oracle9iAS Personalization (OP) is an integrated software application that provides a way for a Web site to customize -- or personalize -- the recommendations it presents to Web site visitors and customers.

Recommendations are personalized for each visitor to the Web site. This has distinct advantages over tailoring recommendations to broad, general market segments. Recommendations are based on a visitor's data and activity such as navigational behavior, ratings, purchases, as well as demographic data.

OP collects the data and uses it to build predictive models that support personalized recommendations of the form "a person who has clicked links x and y and who has demographic characteristics a and b is likely to buy z".

OP incorporates visitor activity into its recommendations in real time -- during the Web visitor's session. For example, OP records a visitor's navigation through the Web site, noting the links that are clicked, etc. All this is data stored for that visitor. The visitor may respond to a Web site's request to rate something, e.g., a book or a movie; the rating becomes part of the data stored for that visitor. All the Web-based behavior for the visitor is saved to a database, where OP uses it to build predictive models. This data can be updated with data collected in subsequent sessions, thereby increasing the accuracy of predictions.

OP works in conjunction with an existing Web application. The Web application asks OP to record certain activities, and the data is saved by OP into a schema.

The Web application asks OP to produce a list of products likely to be purchased by a Web site visitor; a scored list of recommendations compiled from the visitor's current behavior and from data in another schema is passed to the Web application.

A third schema maintains administrative schedules and activities.

What Kind of Data Does OP Collect?

OP collects four kinds of data:

• navigational behavior

• ratings

• purchases

• demographic data

Of these, navigational behavior allows the most flexibility. It can represent anything the Web application wants to consider a hit (e.g., viewing a page, clicking a link/item, etc.).

Visitors to the Web site are of two types: registered visitors (customers) and unregistered visitors (visitors). For customers, OP has both data from a current session and historical data collected over time for a given customer, as well as demographic data. For visitors, there is no historical data, so recommendations are based on current session behavior and demographic data, if available.

How Does OP Collect the Data?

OP collects the data using Java calls provided by the REAPI (Recommendation Engine Application Programming Interface). These calls add information to the RE cache for the specific session, identified by a session ID. The RE finds the correct session ID by looking up one of the following arguments passed in the REAPI calls:

appSessionID -- used by sessionful Web applications (that is, an application that stores an identifier for each session)

customerID -- used by sessionless Web applications (that is, an application that does not store an identifier for each session)

In more detail: The data collected are temporarily stored in a dual buffer cache in the JServ (Java server). Periodically the JServ buffer is flushed and the data are sent to the appropriate RE schema. The session data are then used, combined with historical data, to generate recommendations. Finally, the RE instance periodically flushes the data to the MTR for sessions that have concluded or timed out. The RE only flushes data to the MTR with the data source types specified by its configuration parameters. The data in the MTR is then used to build predictive models for future deployment.

Sessionful and Sessionless Web Applications

Some Web applications are sessionful, i.e., they create a session for each user visit to the Web site. Others are sessionless (stateless), i.e., they do not create sessions.

Regardless of whether the calling Web application is sessionful or sessionless, OP is always sessionful; OP always creates a session internally and maps that session to the Web site's session if there is one.

During the OP session, the Web application can collect data and/or request recommendations.

What Does OP Do with the Data?

OP uses the data to build data mining models. The models predict what the Web site visitor will probably like or buy. The predictions are based on the data collected for that Web site visitor in previous sessions, in the current session, and on demographic information.

The OP Administrator defines a package that contains information needed to build a model or models, as well as information about the database connections. The OP Administrator creates and manages schedules for building the packages, and for deploying the packages to the recommendation engines (REs) that will produce the recommendations. Recommendation engines with the same package are grouped together in recommendation engine farms (RE Farms). These and related terms are defined more fully in the next section.

Models

OP uses one of two algorithms, depending on the type of recommendation requested by the web application. Both algorithms are based on a theorem of Bayes concerning conditional probability. See Appendix A for a description of the algorithms.

OP Components

The user of the OP Administrative UI is assumed to be a DBA or a Java programmer. Getting Started with Oracle9iAS Personalization is designed to introduce a Java programmer or DBA to the basic components and interfaces of OP. These components and interfaces consist of:

• Administrative UI: A browser-based user interface that permits the OP Administrator to build, schedule, and deploy packages, manage RE Farms and REs, and otherwise manage OP. Chapter 2 describes the Administrative UI in detail. The Administrative UI is installed on the system where Oracle9i is installed.

• Recommendation Engine (RE): The RE supports a Web application written in Java for collecting and preprocessing customer and visitor data, and for providing recommendations to those customers and visitors. Access to the RE is provided via the REAPI (Recommendation Engine Application Programming Interface). A given RE may support one or more Java server processes in a Web application.

• Recommendation Engine Farm (RE Farm): A logical grouping of one or more REs that are populated with the same deployable package (data mining model(s)). An RE Farm is generally treated as a single unit for management from the Administrative UI.

• Package: An object created using the Administrative UI. A package contains the information from historical data necessary to make recommendations. A package defines the build settings and other attributes necessary for building data mining models and to schedule model builds.

• Mining Object Repository (MOR): The Oracle database that maintains mining metadata and mining model results as defined by the Oracle9iAS Personalization data mining schema. Via the Administrative UI, the MOR serves as the focus for logging in to the data mining system, logging off, and scheduling OP events.

• Mining Table Repository (MTR): The MTR contains the schema and data to be used for data mining. The MTR has a fixed schema designed to support the building of models that produce recommendations for customers and visitors.

• Recommendation Engine API (REAPI): A collection of Java classes that enable a Web application to collect and preprocess data used to build OP models and to obtain recommendations in real time from OP (that is, to score products for particular customers). OP also includes a Batch RE API that permits users to obtain bulk recommendations in batch, i.e., offline. REAPI and the Batch API are installed on the system where Oracle9iAS is installed.

It is an option during OP installation to populate the MTR with a small amount of sample data. If this option is chosen, an RE demo can be accessed and some recommendations and administrative actions can be tested.

Location of OP Components

OP requires both Oracle9i and the Oracle9i Application Server (Oracle9iAS). Oracle9i and Oracle9iAS are usually installed on different systems.

The following OP components are installed on the system where Oracle9iAS is installed:

• REAPI

• RE Batch API

• REAPI Demo

All other components, including the Administrative UI and OP documentation, are installed on the system where Oracle9i is installed.