Social Summary Generation via Semantic Scoring

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In social networking contexts, users face difficulties in understanding the major points of discussion due to the large number of comments referencing a particular hashtag or web link, as existing technologies do not effectively summarize semantic relationships between communications.

Innovation Solution

A system comprising a storage component, score computation component, and summary generation component that identifies communications with common connectors (such as hashtags or web links), computes similarity and diversity scores, and generates summaries representing the selected communications to highlight dominant themes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If users read through all comments referencing a particular hashtag or web link, then they can understand the major points of discussion, but the time and effort required increases significantly due to the large number of comments

Engineering Contradiction:

ImproveUnderstanding of major discussion pointsVSAvoidTime to read through comments

Core Design Contradiction:

Loss of informationVSLoss of time

Solution Approach 1:

The system extracts only the most representative and informative comments from the large set of all comments. By computing similarity scores and selecting only those comments that best represent the discussion themes, the system extracts the essential information without requiring users to read through every single comment, thus resolving the contradiction between information completeness and time consumption

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system changes the parameter of comment selection from exhaustive (all comments) to selective (representative comments based on similarity scores). By introducing similarity scoring as a selection criterion, the system transforms the approach from reading all comments to reading only those that meet certain similarity thresholds, thereby reducing time while maintaining understanding

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If the system computes similarity scores for all communications, then it can identify representative comments accurately, but the computational complexity and processing time increases

Engineering Contradiction:

ImproveAccuracy in identifying representative commentsVSAvoidComputational complexity of score computation

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the computation process into distinct components: filtering comments with common connectors, computing similarity scores only for filtered subsets, and selecting representative comments. This segmentation allows similarity computation to be performed on smaller, manageable subsets rather than all comments simultaneously, reducing overall computational complexity while maintaining accuracy

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary filtering to identify comments with common connectors before computing similarity scores. This preliminary action reduces the dataset size beforehand, so that the computationally intensive similarity score computation is performed on a smaller subset, thereby reducing computational complexity while preserving measurement precision

Inventive Principle:

Principle #10Preliminary action

3Loss of information

If the system generates a summary of all communications, then it provides comprehensive coverage of discussions, but the summary becomes too lengthy and loses conciseness

Engineering Contradiction:

ImproveComprehensive coverage of discussion topicsVSAvoidConciseness of summary

Core Design Contradiction:

Loss of informationVSProductivity

Solution Approach 1:

The system applies local quality by ensuring that the selected representative comments collectively cover different aspects and themes of the discussion. By carefully selecting comments that represent different local themes while maintaining overall comprehensiveness, the system achieves both comprehensive coverage and conciseness, as only the most informative comments are included rather than all comments

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS10169419B2System and method for generating social summaries

Publication Date: 2019.01.01 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

The described implementations relate to communication platforms that are provided over computer networks. One implementation provides a system that can include a storage component configured to store a plurality of communications having a common connector. The system can also include a score computation component that is configured to compute scores reflecting semantic relationships between individual communications from the plurality of communications. The system can also include a summary generation component that is configured to select one or more of the individual communications, based on the scores, and generate a summary of the common connector. The summary can represent the selected individual communications. The system can also include at least one processor configured to execute one or more of the components.