Providing a script for custom task recommendation criteria

Guidelines for creating scripts in recommendation criteria for an Intelligent Task Recommendation policy.

Customization script return object

Your script should return a JavaScript object in a minimum form as follows:

{ 
  "<task sys_id1>": 
    {
      "rating": <rating1>
    },
  "<task sys_id2>”:
    {
      "rating": <rating2>
    },
  …
}

For filter constraints, set the return objects rating to one to unify the filter constraint result of the recommendation criteria by using the sn_task_recommend.TaskRecommendationUtil.setRatingToOne(your object); method.

To normalize the rating result of ranking criteria, the return JavaScript object should include normalization information.

{           
    "<task <sys_id1>":     
        {
            "rating": <rating1>, 
            "normalizationData": 
                {
                    "numerator": <numerator value1>,
                    "denominator": <denominator value1> 
                }
        },    
    "<task <sys_id2>":     
        {
            "rating": <rating2>, 
            "normalizationData": 
                {
                    "numerator": <numerator value2>,
                    "denominator": <denominator value2> 
                }
        },
    ...
}

For ranking criteria, the return object can contain data for final normalization.

Note: If your scripts include normalization data, you can refer to the default script includes in the predefined recommendation criteria:

• The filtering constraint Exclude tasks agent cannot travel to: sn_fsm_task_rec.TaskRecommendationDistanceRuleProcessor

• The ranking criteria Distance from task: rankTaskOnDistance()

Customized script in recommendation criteria

The following example shows how to write a script for the recommendation criteria.

var customizedScript = <your-script>; 
var customizedResult = customizedScript.<your-method>(); 
ruleResult = TaskRecommendationFSMUtil.parseRuleResult(customizedResult, "<customized-rule>");

The following sample configuration provides a “distance from task” filter constraint.

var distanceRule = new TaskRecommendationDistanceRuleProcessor(args); 
var ruleProcessResult = distanceRule.processRule(user, tasks, timeStart, timeEnd, 'ranking'); 
ruleResult = TaskRecommendationFSMUtil.parseRuleResult(ruleProcessResult, "Distance from task");

Note: Do not replace the task recommendation application keyword ruleResult in the script with any other words. Otherwise, the application will not be able to process the rule execution result.