- Print
- PDF
Version 4.29
November 1, 2024
Version 4.29 now supports an enhanced sentiment analysis model designed for greater accuracy and performance. To maximize the benefits of these enhancements, please update your Sentiment service to Version v11.0 or newer. This update ensures optimized performance and more precise sentiment insights.
If your evaluation forms currently include scoring criteria for "very negative" or "very positive" sentiment, please note that these questions will need adjustment. With the update to the new sentiment service in Version 4.29, results will no longer return "very negative" or "very positive" classifications. We recommend removing these criteria and replacing them with a standard sentiment question to align with the new service capabilities and ensure accurate sentiment assessments
Version 4.29 has been restructured to support the latest redaction service features. Please ensure that your redaction service is updated to Version v1.3.0 or newer to maintain compatibility and leverage the improved redaction capabilities. This update enhances data privacy, accuracy, and customization options. Please note that after that update previously saved redaction settings will be lost. Therefore make sure that related configurations must be reset after updating the application.
Enhancements
Prebuilt Hold Reason Project
An Ask Gen AI project that enables the configuration of prebuilt hold reasons for enhanced tracking during conversations. This simplifies the categorization of call holds, making it easier for users to understand hold purposes. It improves call analysis and reporting accuracy.
Calculating Category Matching Score for User-Defined Categories
A scoring mechanism calculates category matching for user-defined categories, providing insights into how well conversations align with set categories. With that score, the system automatically detects the main category in conversations, streamlining analysis and reporting. It reduces manual categorization work and improves data-driven insights.
Adapting to New Sentiment Service
Adjustments have been made to integrate a new sentiment analysis service. This ensures consistent, accurate sentiment data for user interactions, enhancing the quality of conversation analytics. New sentiment service is capable of analyzing more complicated dialogs and gave more confident responses.
Adding Shortcuts to Email Alerts
Shortcuts for agents, supervisors, and department managers are now available in email alerts, allowing quick access to relevant notifications. This improvement boosts communication and response efficiency within the organization.
Drill-Down Function in Category Distribution Graph
A drill-down feature has been added to the category distribution graph, enabling users to explore specific data points in greater detail. This supports deeper analysis and enhances data visualization.
Improvements
Dynamic Redaction
New Dynamic redaction structure is supported to be able to integrate all entities supported by redaction.
Graceful shutdown for all services
A graceful shutdown process has been implemented for core services, including web, API, identity, and background services. This change supports system stability and minimizes the risk of data loss during service interruptions.
Creating Parallel Structure for Batch Tagging
A parallel processing structure has been implemented for batch tagging, accelerating the tagging process for large data sets. This update increases efficiency in data handling.
Restricting Generative AI Category Items Without Keys
A restriction has been added to prevent users from adding generative AI category items without providing necessary keys. This ensures that categories are configured correctly, maintaining the integrity of AI-driven insights.
One-Click Drill-Down in Conversation Count Graph
A one-click drill-down function has been added to the conversation count graph, enabling users to quickly access detailed data points. This simplifies navigation and facilitates faster analysis.
Elastic APM / OpenTelemetry
Integration with Elastic APM and OpenTelemetry supports enhanced application performance monitoring. This feature enables better tracking and debugging, improving overall system reliability.
Optimization of Product Builds
Optimization techniques have been applied to the product build process, improving efficiency and reducing build times. This results in a more streamlined deployment process.
Minor UI Enhancements
Several minor improvements have been implemented to enhance the user interface. These enhancements make the interface more intuitive and user-friendly, contributing to a better overall user experience.
Minor Bug Fixes
Various minor bugs have been addressed to improve overall system stability and performance. These fixes result in a smoother and more reliable user experience.