Telecom Interaction Tagging with PMI-Based Transcript-Summary Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing tagging systems for telecommunication interactions are inefficient and inaccurate, particularly when dealing with vague, lengthy, or noisy transcripts, leading to excessive computing resource consumption.
Innovation Solution
A system that utilizes both transcripts and agent-prepared summaries to generate indices of terms, calculates association scores using pointwise mutual information, and extracts candidate key phrases for accurate tagging, conserving computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional tagging systems process telecommunication transcripts, then tagging functionality is provided, but computing resource consumption is excessive and accuracy is poor
Solution Approach 1:
The system performs preliminary actions by generating term indices and association score datasets from historical transcripts and summaries before actual tagging occurs. This pre-processing creates reusable resources that accelerate subsequent tagging operations while maintaining high accuracy, thereby reducing real-time computing resource consumption.
Solution Approach 2:
The patent introduces intermediary elements including term indices, association score datasets, and candidate key phrase lists that mediate between the raw transcript and final tags. These intermediaries structure and pre-process information, enabling more efficient and accurate tagging without requiring excessive computing resources during the actual tagging operation.
2Measurement precision
If comprehensive term indexing is performed on transcripts and summaries, then tagging accuracy improves, but processing time increases
Solution Approach 1:
Term indexing and association score calculation are performed as preliminary actions during offline or batch processing phases. The generated indices and datasets are stored for reuse, allowing rapid tagging of new transcripts without repeating the expensive indexing operations, thus achieving high accuracy without proportional increases in processing time.
Solution Approach 2:
The system performs partial indexing by focusing on generating term indices and association scores only for the most relevant terms and phrases identified through the association score methodology. This selective approach achieves sufficient tagging accuracy without the need to process and index every possible term, thereby reducing overall processing time.
Data Source
AI summary
In some implementations, a device may obtain a data set that indicates scores for first terms based on association scores between the first terms and second terms. The association scores may be based on occurrences of the first terms in electronic documents that provide transcripts relating to telecommunication interactions, and occurrences of the second terms in electronic records that provide summaries relating to the telecommunication interactions. The device may process an electronic document, that provides a transcript relating to a new telecommunication interaction involving an agent device and a user device, to extract candidate key phrases from the transcript. The device may determine a relevance score for each candidate key phrase based on the score(s), of the data set, associated with the first term(s) in the candidate key phrase. The device may generate tag(s) for the new telecommunication interaction based candidate key phrase(s) associated with highest relevance score(s).