IVR Intent Prediction Using User History and 5G Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Interactive Voice Response (IVR) systems often present users with complex static menus, leading to user confusion and frustration, as they struggle to navigate through multiple selections without remembering previous choices.
Innovation Solution
An IVR system that predicts user intent by analyzing interaction history, billing status, and device data, using machine learning to provide dynamic menus and engage appropriate software applications or bots to address user needs, even on non-5G devices, and offers visual content and video conferencing for enhanced interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If static menus are used in IVR systems, then the system structure is simple and easy to implement, but user experience deteriorates due to confusion and frustration when navigating complex options
Solution Approach 1:
The patent applies dynamics by transitioning from static menus to dynamic menus that automatically adapt based on user characteristics and interaction history. The system analyzes user data and adjusts menu structure in real-time, making the menu flexible and personalized rather than fixed, thereby improving ease of operation while maintaining manageable system complexity through automated adaptation
Solution Approach 2:
The system implements self-service by automatically analyzing user interaction history and billing status to predict user intent without requiring manual configuration. The IVR system serves itself by gathering data, processing it through machine learning models, and automatically generating personalized menu structures, eliminating the need for manual menu customization for each user
2Adaptability or versatility
If multiple selections are required in IVR menus, then service coverage is comprehensive, but user memory capacity is exceeded leading to loss of selections and increased frustration
Solution Approach 1:
The patent applies preliminary action by predicting user intent before the user actually makes selections. The system analyzes interaction history and billing status in advance to anticipate what the user needs, presenting only relevant options beforehand. This prevents users from having to remember multiple selections by pre-filtering and pre-presenting the most likely needed services
Solution Approach 2:
The system implements local quality by tailoring the menu structure to individual user needs rather than presenting a uniform comprehensive menu to everyone. Each user receives a customized menu focused on their specific requirements based on their interaction history and billing status, making the service coverage relevant and manageable for each individual
3Ease of manufacture
If traditional IVR systems are used, then device requirements are minimal, but bandwidth efficiency is poor limiting the delivery of rich visual content and video conferencing
Solution Approach 1:
The patent applies parameter changes by detecting whether the user device is a 5G capable device and adjusting the communication parameters accordingly. When 5G capability is detected, the system enables enhanced services with higher bandwidth utilization for visual content and video conferencing. When not detected, the system automatically falls back to traditional IVR modes, thus adapting the bandwidth requirements to match device capabilities
Data Source
AI summary
Disclosed here is a method to determine a user intent when a user device initiates an interactive voice response (IVR) call with a wireless telecommunication network. A processor can detect the IVR call initiated with the network and determine whether the user device is a member of the network. Upon determining that the user device is a member of the network, the processor can obtain user history including interaction history between the user and the network. Based on the user history, the processor can predict the user intent when the user initiates the IVR call. The processor can detect whether user device is a 5G capable device. Upon the determining that the device is 5G capable and based on the predicted user intent, the processor can suggest to the user an application configured to execute on the user device and configured to address the predicted user intent.