New features and changes in HCL Opportunity Detect 10.0

Trend components (parity)

Trend components detect changes in activity measured over defined periods of time. Before you can create a trend component, your trigger system must include a Container or Select component that the trend can use as a data source.

Three components track trends.

Trend component

Detects simple trends. For example, the average rolling monthly balance has been increasing or decreasing by 10% over the last three months.

Spike component

Detects an unusually large change in activity. For example, a deposit is 50% larger or smaller than any other deposit made in the last 3 months.

Exceeded Standard Deviation (ESD) component

Detects activity outside of the standard variance for a specified time period. For example, a monthly balance that is greater than or less than the average monthly balance for the past 12 months by two standard deviations.

Sequence pattern (parity)

With the Sequence Pattern component, you can specify a series of incoming events, and set the number of times the complete sequence of events must occur before the component fires.

Like operator (parity)

When you create a Boolean expression in a component and you select the Like operator, you can use a regular expression to set the criteria for comparison.

The regular expression must conform to the Streams Processing Language requirements.

Date functions (parity)

The following features have been added.

• You can subtract dates in Math expressions. For example, you can subtract a date derived from a data source from a date constant such as the beginning of the current month.
• When you define a time span, you have the following options for the Type field in the Select Period window.
  • Constant
  • Container
  • Date Constant
  • Data Source
  • Math Expression
  • Select

  This expands your options for the way the number of time units is calculated.

• When you define a time span, you have the following options for the Value field in the Select Period window. These time constants are based on the current transaction, and they return an integer, as shown below.

  Table 1. Date constants that return an integer

  Date constant	Value
  Current Second	1 - 60
  Current Minute	1 - 60
  Current Hour	1 - 24
  Current Day of the Month	1 - 31
  Current Day of the Week	1 - 7 Sunday = 1
  Current Week	January 1 = day 1 of week 1 In a leap year, December 30 is in week 53
  Current Month	1 - 12
  Current Quarter	1 - 4
  Current Year	Four digits

GroupBy function (parity)

The Group By function is available in Select and Container components. This function groups together all the records that have identical values in this field and returns one or more groups of rows. The Group By function can be applied only to a single field in a component.

Additional delimiters for batch files (parity)

You select from a list of the following available delimiters when defining a file connector.

• pipe (vertical line: |)
• comma
• semi-colon
• tab

Export / import utility

The export/import utility transfers Opportunity Detect workspace logic from one environment to another.

The export/import utility operates at the workspace level. You can export all workspaces or specify the workspaces you want to export.

The utility creates an XML file that contains a specification of the workspace logic and all of its dependencies. The export includes all the defined data sources, audience level codes, and named value lists regardless of their usage in the exported workspace.

After export, you can transfer the XML file to the target environment and import the logic. After an export/import cycle, the logic transferred is functionally equivalent in both environments.

Monitoring utility

Opportunity Detect provides a monitoring tool for deployments running on Streams. The monitoring tool runs as a JConsole service. You can view details in real time at three levels: deployment, Streams server, and Opportunity Detect engine. You can also monitor the health of the Real Time Web Connector service and the Remote Control Service.

Any change in deployment status from healthy to unhealthy triggers a notification to Opportunity Detect Opportunity Detect users, who can subscribe to receive these notifications in their inbox and also in email, if email notifications have been configured. Administrators can also create notifications that appear in the tool, using the JConsole notifications feature.

If you plan to use the monitoring tool, the names of your deployment configurations must be unique across your entire Opportunity Detect installation. If you have used the same name for multiple deployment configurations, change the names to ensure that they are unique

Delayed insert

You can select a Delayed insert option in Container Manipulator components. When you choose the Delayed insert option, the system adds the values to the contents of the Container only after all logic in the trigger system has finished processing the transaction.

Real time file connector

Real time file connectors enable Opportunity Detect to connect to transaction data that is in fixed width file format and that is updated frequently. Opportunity Detect reads these files in real time. You configure your automation system to place the properly formatted file in an input directory that you specify when you create the connector in the Opportunity Detect user interface.

When you create the real time file connector you have the option to apply a Bloom filter to eliminate duplicate data based on fields that you specify.

After Opportunity Detect consumes the data, the files are automatically moved to another directory that you specify for precessed files.

On demand access for profile data

By default, Opportunity Detect caches profile data, which provides optimal performance. However, for very large profile data sets, you have the option to switch to on demand access for profile data. This eliminates the out of memory issues that caching large amounts of profile data can cause.

To enable on demand access to profile data, go to the Settings > Configuration page, open the IBM Opportunity Detect and Interact Advanced Patterns | System | Processing Options category, and set the Cache profile records property to false.

Support for Rabbit MQ in queue connectors

Queue data source connectors can now work with Rabbit MQ in addition to the previously supported Active MQ and WebSphere MQ servers.

New option during installation to create a system table data source connection in the web application server

You can allow the installer to create the connection to the Opportunity Detect system tables in the web application server. To enable automatic data source creation during installation, on the Datasource Creation panel, select the Create Opportunity Detect Datasource check box and supply information about your application server.

• The installer creates the data source using campaignPartition1DS OpDetectionDS collaborateds plands UnicaPlatformDS as the JNDI name.
• The installer creates the data source connection for one partition only. If you have multiple partitions, you must create additional connections manually.
• Note that, if you are using WebLogic, you must add the JDBC driver to your web application server classpath manually even if you allow the installer to create the data source. The installer does this automatically for WebSphere.
• The installer adds the JDBC driver to your web application server classpath automatically.

Context-sensitive help for Configuration properties

When you view configuration properties under Settings > Configuration, you can choose Help > Help for this page to display context-sensitive help. For example, if you request help for the Campaign | caching page, you see a help topic that specifically describes all of the caching options. You must have an internet connection to use this feature.

Upgrade paths

Because version 10.0 is a major release, you can upgrade directly from 8.6.x, 9.0.x, or 9.1.x in a single step. For instructions, see your product's Upgrade Guide.