OVERVIEW
Virtual Assistants
Kore.ai Platform
Key Concepts
Natural Language Processing (NLP)
Accessing Platform
VIRTUAL ASSISTANTS
Virtual Assistant Builder
Virtual Assistant Types
Getting Started
Creating a Simple Bot
SKILLS
Storyboard
Dialog Task
Introduction
Dialog Builder (New)
Dialog Builder (Legacy)
User Intent Node
Dialog Node
Entity Node
Supported Entity Types
Composite Entities
Supported Colors
Supported Company Names
Form Node
Logic Node
Message Nodes
Confirmation Nodes
Bot Action Node
Service Node
Custom Authentication
2-way SSL for Service nodes
Script Node
Agent Transfer Node
WebHook Node
Grouping Nodes
Connections & Transitions
Manage Dialogs
User Prompts
Knowledge Graph
Terminology
Building
Generation
Importing and Exporting
Analysis
Knowledge Extraction
Build
Alert Tasks
Introduction
Ignore Words and Field Memory
How to Schedule a Smart Alert
Small Talk
Digital Views
Overview
Configuring Digital Views
Digital Forms
Overview
How to Configure Digital Forms
NATURAL LANGUAGE
Overview
Machine Learning
Introduction
Model Validation
Fundamental Meaning
Introduction
NLP Guidelines
Knowledge Graph
Traits
Introduction
How to Use Traits
Ranking and Resolver
Advanced NLP Configurations
INTELLIGENCE
Overview
Context Management
Overview
Session and Context Variables
Context Object
How to Manage Context Switching
Manage Interruptions
Dialog Management
Sub-Intents & Follow-up Intents
Amend Entity
Multi-Intent Detection
Sentiment Management
Tone Analysis
Sentiment Management
Event Based Bot Actions
Default Conversations
Default Standard Responses
TEST & DEBUG
Talk to Bot
Utterance Testing
Batch Testing
Conversation Testing
CHANNELS
PUBLISH
ANALYZE
Overview
Dashboard
Custom Dashboard
Overview
How to Create Custom Dashboard
Conversation Flows
NLP Metrics
ADVANCED TOPICS
Universal Bots
Overview
Defining
Creating
Training
Customizing
Enabling Languages
Store
Smart Bots
Defining
koreUtil Libraries
SETTINGS
Authorization
Language Management
PII Settings
Variables
Functions
IVR Integration
General Settings
Management
Import & Export
Delete
Versioning
Collaborative Development
PLAN & USAGE
Overview
Usage Plans
Support Plans
Invoices
API GUIDE
API Overview
API List
API Collection
SDKs
SDK Overview
SDK Security
SDK App Registration
Web SDK Tutorial
Message Formatting and Templates
Mobile SDK Push Notification
Widget SDK Tutorial
Widget SDK – Message Formatting and Templates
Web Socket Connect & RTM
Using the BotKit SDK
Installing
Configuring
Events
Functions
BotKit SDK Tutorial – Agent Transfer
BotKit SDK Tutorial – Flight Search Sample Bot
Using an External NLP Engine
ADMINISTRATION
HOW TOs
Creating a Simple Bot
Creating a Banking Bot
Context Switching
Using Traits
Schedule a Smart Alert
Configure UI Forms
Add Form Data into Data Tables
Configuring Digital Views
Add Data to Data Tables
Update Data in Data Tables
Custom Dashboard
Custom Tags to filter Bot Metrics
Patterns for Intents & Entities
Build Knowledge Graph
Global Variables
Content Variables
Using Bot Functions
Configure Agent Transfer
Update Balance Task
Transfer Funds Task
RELEASE NOTES
  1. Home
  2. Docs
  3. Virtual Assistants
  4. Natural Language
  5. Traits

Traits

In natural conversations, it is very common that a user provides background/relevant information while describing a specific scenario. Traits are specific entities, attributes, or details that the users express in their conversations. The utterance may not directly convey any specific intent, but the traits present in the utterance are used in driving the intent detection and bot conversation flows.

For example, the utterance my card is being rejected and am on a business trip expresses two traits card decline and emergency. In this scenario, the utterance does not convey any direct intent, or at best it is used to trigger the unblock card flow. However, the presence of emergency trait is used to directly assign the conversation to a human agent.

Traits feature of the Bots platform is aimed at identifying such characteristics present in user utterances and use them for intent detection and in customizing the bot definition using these characteristics.

Use Case

Book a Flight bot might have an added requirement to book a flight based on the cost preference.

User utterance: I am looking for a low-cost option to London must result in ordering the available flights and picking the lowest-priced ticket.

This can be achieved by:

  • Adding a Trait Type called Travel Class with Trait Economy trained with the utterance low cost.
  • Adding a Rule for book flight to be triggered in the presence of Economy Trait.
  • Add transition condition in case Trait Economy is present in the context.

Configuration

Configuring Traits involves:

  • Trait Definition
  • Trait Association Rules
  • Trait Detection

Trait Definition

To define Traits, under the Build top menu option, from the left menu, click Natural Language –> Training and select the Traits tab.

Following are the key features to be considered while defining Traits:

  1. Trait Type is a collection of related traits like Travel Class in the above example.
  2. Trait Type can be ML Based or Pattern Based. Each trait of a trait type can be trained using words, phrases, utterances, or patterns based on the type. Manage Trait Type allows you to define the training configuration. See below for ML-based trait configuration.
  3. A Trait Type can have one or more Traits.
  4. Traits names should be unique in a group. But traits with the same name can be present in multiple groups.
    • For ML-based Traits, you can define the words, phrases, or utterances that identify the trait. One trait per trait type is detected for ML-based trait types.
    • For Pattern-based Traits, you can define the patterns associated with the given trait. There is a possibility of multiple traits getting detected for pattern-based trait types. Ordering of Traits within the Trait Type signifies the importance of a trait in a trait type and detects only one trait.
  5. Once added, Train the bot for the Traits to be detected from user utterances.

 

Notes:

  1. You can add language-specific traits in the case of multi-lingual bots.
  2. When a trait name is modified, ensure that all the rules defined using that trait are corrected. This has to be done manually, the platform will not take care of it.
  3. The trait name must be unique in a group.
  4. Traits with the same name can be present in multiple groups, but distinguishing them in trait rules or trait detection results is difficult.

Traits – ML Model

When choosing to train traits using the ML model, by default, the n-gram model is used. An n-gram is the contiguous sequence of words used from training sentences to train the model. But this might not be effective when the corpus is very less or when the training sentences, in general, contain fewer words.

From v8.0 of the platform, an option is included to skip or use the n-gram model. Further, the option to parameterize the n-gram algorithm is included.

  • When the n-gram option is selected, you can configure the n-gram Sequence Length by setting the maximum value of the n-gram. It is set to 1 by default and can be configured to any integer value between 1 and 5.
  • When skip-gram is selected, you can configure
    • Sequence Length specifying the number of words to be included in a non-consecutive sequence. It is set to 2 by default and it can take any integer value between 2 and 4.
    • Maximum Skip Distance for the number of words that can be skipped to form a non-consecutive sequence of words. This value is set to 1 by default and can take any integer value from 1 to 3.

NOTE: While the settings are same for all languages (in case of multilingual bot), for some languages like Chinese and Korean sequence of characters form grams and for other (Latin-based) languages are word grams.

Trait Association Rules

Trait Rules define Dialog Execution and Knowledge Graph Intent detection.

Dialog Execution

Intent detection or Dialog execution is achieved using traits, along with the ML utterances and patterns. To achieve this, intent must be associated with the required traits by adding Rules.

There are multiple ways to add rules:

  1. From the Traits section using the Add New Rule link.
  2. From the Intent Node using the Rules section under the NLP Properties.
  3. From the Rules tab for a given Intent.

Each rule can have one or more conditions with AND as the operator. Multiple trait rules can be defined for a given intent and the intent is considered as a definite match if any one of the rules matches.

Knowledge Graph Intents

Knowledge Graph can also be part of the discovery process using Traits. For this, each term or node can be associated with a trait. A given term can be associated with a single Trait.

Trait Detection

Only one trait from a group (trait type) will be detected and is considered as a definite match.

Traits detected are included in the context object. The context is populated with unique traits identified (without reference to trait type). This information can be used in:

  • Intent identification
  • Dialog transition
  • Entity population
  • bot definitions

Batch Testing reports also include information about traits detected as do the Find Intent API.

Intent Detection

The Ranking and Resolver gets input from the three NL engines and Traits to analyze and come up with the possible/definitive matches.

  • The intent is considered as a definite match only if all the traits (one in the case of Knowledge Graph) present in a trait rule are detected.
  • NL Analysis includes information on traits detected and the NLP Flow shows the information about traits detected.

Dialog Transition

Conversation Flow is controlled using Traits. For a Dialog, Connection Rules are defined using the Trait Context. This is done from the Connection tab under the Properties Panel for the Dialog.

The Traits Context is accessed using context.traits. It returns an array of all traits matching the intent, hence the condition to be used is contains.

Menu
Kore.ai Named a Leader in 2022 Gartner® Magic Quadrant™ for Enterprise Conversational AI PlatformsGet the Report