1. Docs
  2. Virtual Assistants
  3. What's New
  4. 8.0.x Release Details

8.0.x Release Details

This page gives full details of the various platform releases and the corresponding features that were updated/enhanced in version 8.0.x of the platform.

v8.0.08 Jan 9th 2021

Minor Update

This update was mainly aimed at addressing some minor bugs.
Following is a list of few updates.

Feature Enhancement
Natural Language
NLP Configurations
  • Network Type Transformers has been introduced to support Universal Sentence encoder, refer here for details.
  • ML Intent detection behavior when multiple sentences are present in user utterance can be configured, refer here for details.
  • Rescoring behavior in presence of possible intents from multiple engines can be configured, refer here for details.

v8.0.07 Dec 26th 2020

Minor Update

This update was mainly aimed at addressing some minor bugs.
Following is a list of few updates.

Feature Enhancement
Natural Language
Entity Processing
  • The number of retries allowed for the user input can be customized and the bot behavior when a user exceeds the retry limit can also be defined, refer here for details
  • The order in which the prompts are presented to the user in the event of a wrong entry for an entity value can be customized, refer here for details
  • Default values can be set for Hidden and Optional entities, refer here for details
  • The entities identified from user utterances can be replaced with the respective entity placeholders for better intent detection, refer here for details
  • Passing of values for Composite Entities across Dialog Tasks has been enhanced and simplified, refer here for details
NL Behavior
  • NLP engine can now recognize emojis when present in the user utterance. You can include these in the concept definition and give a better user experience, for example, a thumbs-up emoticon from the user can be considered as an affirmative and a frown emoticon can be considered as an anger sentiment and appropriate action can be taken.
Miscellaneous
  • API to modify the NLP Thresholds and Configurations and Advanced Configurations has been made public, refer here for details.

v8.0.06 Dec 12th 2020

Minor Update

This update was mainly aimed at addressing some minor bugs.
Following is a list of few updates.

Feature Enhancement
Natural Language
Negation of Intents The negation of trained intents will be ignored by the platform, for example, “I don’t want to transfer funds” will not trigger the intent “Transfer Funds”.
Support to this feature has been extended to Spanish language.

v8.0.05 Dec 6th 2020

Minor Update

This update was mainly aimed at addressing some minor bugs.

v8.0.04 Nov 28th 2020

Minor Update

This update was mainly aimed at addressing some minor bugs.
Following is a list of few updates.

Feature Enhancement
Bot Builder
Bot Import & Export The enableNegativePatterns flag will now be part of ‘NLP Settings’ under ‘NLP Data’ in the exported and imported bot definitions. This flag is used to check if the Negative Patterns are enabled/disabled in the Bot.
Prior to this patch, this flag was part of ‘NLP Settings’ under ‘NLP Data’ in the imported bot definitions and ‘Bot Setting’ under ‘Setting’ in the exported bot definitions. This change is aimed at streamlining the functionality and might result in the failure to the incremental import of a bot from an export file generated before v8.0.04.

v8.0.03 Nov 21st 2020

Minor Update

This update was mainly aimed at addressing some minor bugs.

v8.0.02 Nov 14th 2020

Minor Update

This update was mainly aimed at addressing some minor bugs.
Following is a list of few updates.

Feature Enhancement
Platform
Multi-language support Bot Builder platform can now be accessed in English and Japanese languages. Know more.
Natural Language
k-fold cross validation The kFold report needs to be generated explicitly using the Generate/Regenerate button. Know more.

v8.0.01 Nov 7th 2020

Minor Update

This update was mainly aimed at addressing some minor bugs.
Following is a list of few updates.

Feature Enhancement
Natural Language
k-fold cross validation A minimum of 250 ML utterances are needed for effective performance of k-fold validation. This number can be customized via ML config for on-prem installations. This was implemented to address the issue of high CPU utilization of ML server. Know more.

v8.0.00 Oct 31st 2020

Major Release

Following is a list of feature enhancements addressed in this release.

Feature Enhancement
Bot Builder
Storyboard The storyboard experience has been enhanced with the ability to link scenes from another story; ability to access version history and revert to a previous version; enhanced comment behavior for better collaboration; ability to add tags to Notes, change the background image of the Preview page, and support for image template and more. Learn More.
Small Talk Small Talk improvements include the identification of Small Talk intents in batch testing (know more) and access to the contextual information from the user’s utterance  (know more).
Bot Import Bot Import functionality has been enhanced to completely override the bot definition in the in-development copy with the definition present in the import file with an option to auto-publish the bot post successful bot import. Learn more.
Bot Versioning Bot Versions can now be exported. Learn more.
Universal Bot Universal Bot enhancements include event handling; Fallback bots; Inclusive bots; training utterances per linked bot; in-context linked bot’s information and more. Learn More.
Variable Management To facilitate the loading of only relevant variables while executing specific components for improving the bot’s performance and response times. Learn more.
Miscellaneous
  • Users can be allowed to set an Alert on any day of the week in addition to the current daily, weekends, and weekdays. Learn more.
  • Synchronous webhook channel support has been extended for Digital forms.
Feature Enhancement
Natural Language
NL enhancements include validation of the machine learning model using K-fold cross-validation metrics (know more); training of intent and trait models using the skip-gram algorithm (know more); additional configurations for the NL engines (know more).
Knowledge Graph Document Annotation Tool (know more); context output and input pre-conditions for Organizer nodes (know more); the Knowledge Graph Analyzer to support multi-word synonyms; inclusion of Knowledge Graph’s Training status in the Status Docker (know more); Knowledge Graph related API to fetch FAQs and APIs related to Knowledge Extraction.
Entity Processing Inclusion of confidence scores for entities identified in testing results – utterance and batch (for CRF model) (know more); Entity rules for additional processing hints and validations (know more); Composite Entity enhancement to handle JSON objects as the input (know more); Amend Entity support for Hidden entities; entity identification in Batch Tests and Utterance Testing when the entity order is not provided.
NL Behavior
  • Improvements to PoS (part-of-speech) tagging by the Fundamental Meaning engine would ensure that words are recognized correctly as nouns and verbs and thus intent detection would be more accurate. Note this change might result in a slight deviation in the success percentage of the existing bot training test cases.
  • NL Enhancements: Chat script, the force behind NL processing has been upgraded. This will ensure a better bot behavior in processing user intent and utterances. Following are areas where its impact might be felt:
  • Some minor changes to the underlying grammar rules for English POS tagging may result in some slight differences in the NL analysis like refined demarcation between nouns and verbs in user utterances.
  • Interjections, like “hello” at the start of an user utterance, were being split into their own sentence and replaced with a concept like ~emohello. Ex: “Hello, I need to transfer money” was being split into 2 different sentences “~emohello” and “I need to transfer money”. It will not be split now and would be matched with the intent “transfer money”. The interjections are still marked with the concepts and so patterns that test for them will still work without changes.
  • All the yes/no synonyms, including the internal ones, are considered at the same time and the earliest choice taken instead of considering the yes choice exclusively first and then the no synonyms. This means that something like “of course not” is now truly considered as NO. The yes/no synonyms can now be concepts and patterns as well.
  • There is a change in the scoring process when there are 3 or more sentences in the utterance. Everything is treated as a possible match rather than a definite (because a single definite in one sentence is not significant enough). Intents accumulate a bonus based on how many sentences they are matched in instead of a big bonus for the first sentence and decreasing after that.
  • Support for new pattern syntax has been added. These include:
    • word1~concept2; ~concept1~concept2
    • ~intent
    • $currentEntity
    • *~n

    Learn more

Feature Enhancement
Bot Analytics
Containment Metrics This is a one-stop place for sessions trend with the breakup of self-service, drop-off, and agent sessions along with insights into tasks and intents used in these sessions. Learn more.
Conversation Sessions
  • Proactively close conversation sessions after a certain idle time. Learn More.
  • Various analytics features and relevant APIs allow you to filter the data for interactive and non-interactive sessions.
Conversation History View complete user information while reviewing the Conversation History.
Miscellaneous
  • Filter the Chat History using the user identity from the Channels.
  • Filter and download Changelogs in Bot Builder and Audit Logs in Admin Console.
  • Custom Meta Tags can now be added directly from BotKit SDK, Web / Mobile SDK, IVR, and Webhook channels.
  • Introducing koreUtil.getSessionId function to get the sessionId, a unique Conversation Session identifier, of the current ongoing conversation. This identifier can be used in generating custom analytics.
  • Get Analytics API to extract the metrics data related to the intent detection and task performance. Know more.
  • User Engagement Trend API to provide the number of users who have interacted with the bot in a  given period across different time intervals. Know more.
Feature Enhancement
Language Support Updates
  • Korean – Platform now maintains a list of words that should not be split into canonical forms.
  • Italian – Dependency Parsing Model support
  • German – Support for Compound words splitting
  • Japanese, Russian and Kazakh – Enhancements to sentiment identification and autocorrection 
Channel Enablement
  • New channels

    • Skype for Business On-Premise channel. Know more.
    • IVR – AudioCodes voice channel. Know more.
  • Ability to request the bot response format of a different channel than that of the current conversation channel. For example, request the platform to return the response in the WebSDK channel format when the user is interacting from the Webhook channel
Enterprise Features
  • Ability to customize the customer Data retention period. Know more.
  • AD Sync support extended to OUD (Oracle Unified Directory). 
  • Availability of the JWT Token generator tool on Kore.ai Developer Portal. Know more.
  • Support expiry of JTI for BotKit JWT. Know more.
Security & Performance TLS support for Redis; Encryption of RabbitMQ payload; MySQL, Node.js have been upgraded.
메뉴