Conversational Branch Prediction Using Section-Level Website Embeddings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Contemporary automated conversational systems lack section-level accuracy in identifying user interests on websites, failing to provide relevant information for visitors interested in specific topics beyond page-level organization, often providing insufficient or inaccurate data.
Innovation Solution
A conversational branch data prediction system that utilizes machine-learning techniques to generate vector embeddings for website sections, identify subsets with particular topics, and predict conversational data based on user interactions across multiple webpages, enhancing accuracy and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If contemporary automated conversational systems use page-level organization to identify user interests, then the system structure remains simple, but the accuracy of identifying specific topic interests deteriorates
Solution Approach 1:
The patent segments webpages into sections and further into sentences, creating a hierarchical structure where each segment can be independently analyzed. This segmentation enables the system to identify user interests at the sentence level rather than just page level, improving measurement precision while maintaining manageable complexity through structured organization
Solution Approach 2:
The patent transitions from page-level analysis to section-level and sentence-level analysis, adding dimensional granularity to the identification process. By introducing these additional dimensions of organization, the system achieves more precise topic identification without fundamentally redesigning the entire architecture
2Reliability
If contemporary automated conversational systems provide page-level information, then the information structure remains simple, but the relevance of information to user interests deteriorates
Solution Approach 1:
The patent segments information into pages, sections, and sentences, allowing the system to provide more relevant information by identifying which specific segments users are interested in. This segmentation enables the conversational system to reliably match user interests with appropriate information segments rather than providing generic page-level information
Solution Approach 2:
The patent applies local quality by analyzing and weighting different sections and sentences differently based on user interaction patterns. The system identifies which local segments (sentences within sections within pages) are most relevant to user interests and prioritizes those in the conversational response, improving reliability without requiring complete redesign of the information structure
3Measurement precision
If the system analyzes section-level interactions across multiple webpages, then the accuracy of topic identification improves, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary action by pre-processing and organizing webpage content into sections and sentences before user interaction occurs. The system pre-identifies and stores section-level information about topics and relationships, so that during actual user interaction, the system can quickly retrieve and match relevant information without performing complex analysis in real-time
Solution Approach 2:
By segmenting the analysis into pre-computed section-level units, the system reduces the computational burden during real-time interaction. The segmentation allows the system to work with pre-processed data structures that require less computational resources to process compared to analyzing complete pages or performing de novo analysis during user interactions
Data Source
AI summary
A conversational branch data prediction system predicts conversational branch data that can be used in automated conversational services (e.g., “chatbots”). The conversational branch data prediction system predicts the conversational branch data based on interactions with multiple sections across multiple webpages of a website, such as sections that include particular portions of text on webpages and omit additional portions of the text. For each section, the conversational branch data prediction system determines a vector embedding of text data in the section and a topic. Based on event metrics data, the conversational branch data prediction system identifies interactions with a particular subset of the sections having a particular topic. A trained machine-learning dialogue model identifies conversational text data correlated with vector embeddings associated with the particular subset of sections. The conversational branch data prediction system provides the identified conversational text data to an additional computing system.