Agent Intelligence (AI) Micro-Certifications

Solution Definition

A configuration record that specifies how to train a predictive model

Predictive Model

A trained machine-learning algorithm used to predict field values during record creation

Due Note: 1

AI Prediction capability can be extended to other tables in ServiceNow.

Process Flow: 2 functional areas for the Agent Intelligence process

Administrator and the Platform

A Solution Definition creates a Solution

An algorithm used to predict a target field value given one or more input field values

Characteristics of a good training set:

Complete Accurate Sufficient

Benefits of AI

Eliminates Bottlenecks Improves Productivity Greater Customer Satisfaction

Review Solution Statistics:

Estimated Solution Precision Estimated Solution Coverage Class Confidence

Solution Definition Steps

Identify the ServiceNow table Identify the training data set Identify input field (Short Description, Location, CI) Identify output fields Determine training frequency

Administrator

Identify the data set you want to use to train the solution, along with the inputs and outputs. After identifying the training data set, inputs and outputs, you will put this into a solution definition. Remember, you can also determine the training frequency and language support within the solution definition. After the solution definition is created, you will submit a training request. Your request is then sent to one of our training servers where machine learning will process your data and create a solution. Please note, after the server is finished processing your request, all data is removed. Once you receive the solution, you will need to test it and determine if any changes are needed. If so, tune and retrain. Otherwise, you can implement the solution.

Ways to train a solution

Inactive Solutions installed with AI New Solutions Active Solutions that are modified

Failed Predictions:

Locate failed predictions by navigating to: - ITSM: Incident > Open - Unassigned - CSM: Customer Service > Cases > Cases Skipped by Agent Intelligence

Benefits of Machine Learning

Prevent Future Issues Categorize & Route Work Benchmark Against Peers Predict Performance

AI Solutions

ServiceNow offers out-of-the box solutions for: - ITSM - CSM - HR

Precision

The aggregate percentage of correct predictions.

Coverage

The aggregate percentage of records that receive the prediction.

Classes

The output field values that the model uses to make predictions. Each class is an output field value with a list of possible precisions, coverage, and distribution metrics to choose from.

Class Distribution

The percentage of records in the table that have the output field value.

Due Note: 3

a class record is created for each output field value. A list of top words is also available per output value.

Solution Definition

a configuration record that specifies how to train a predictive model

Training Data Set

a set of example records for the AI Solution to "learn" the right way of ticket assignment, classification, and prioritization.

Machine Learning

a subset of Artifical Intelligence that often uses statistical techniques to give computers the ability to "learn" (i.e., progressively improve performance on a specific task) with data, without being explicitly programmed

Solution

an algorithm used to predict a target field value given one or more input field values.

Agent Intelligence (AI)

an artificial intelligence solution embedded in the ServiceNow platform

Prediction

An informed guess based on past observations (historical data)

Due Note: 2

Some customers have gamified cleanup of data for their employees. For initial training, 30,000 records (minimum) should be provided for the best results.

Preparation (Steps)

1. Improve Data Quality 2. Understand and Communicate your KPIs 3. Define Inputs and Outputs 4.Foster Organizational Change