Social Influence Scoring for Mobile Network Loyalty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions for social network management in mobile communication networks lack integration of social graph discovery, marketing, and churn management, failing to generate context-neutral influencer scores that can optimize business objectives such as churn management and campaign target discovery, and do not combine connectivity structures from diverse sources with mobile user interaction data.
Innovation Solution
A system and method that automatically derive social influence scores from user interaction graphs in mobile service usage, considering various aspects of user behavior and interactions, by integrating data from online social networks and professional networks to determine an underlying social network structure and generate influencer scores for loyalty and customer experience management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional marketing solutions focusing on observable user properties and KPIs are used, then implementation is simple, but the ability to optimize diverse business objectives such as churn management and campaign target discovery is insufficient
Solution Approach 1:
The patent combines social graph discovery, marketing analytics, and churn management systems into a unified platform that processes user interaction data from multiple sources (mobile networks, online social networks, email networks) to generate comprehensive influencer scores for optimizing diverse business objectives
Solution Approach 2:
The system generates context-neutral influencer scores that can be universally applied to multiple business objectives including churn management, campaign target discovery, loyalty management, and quality of service management, making the system multi-functional and adaptable
2Adaptability or versatility
If social graph discovery, marketing, and churn management are handled in separate manners, then each function can be implemented independently, but integration between these aspects is lacking and overall effectiveness is reduced
Solution Approach 1:
The patent integrates social graph discovery, marketing analytics, and churn management into a single automated system that discovers social networks from user interaction graphs and generates influencer scores that feed into all three functional areas simultaneously
Solution Approach 2:
The system automatically discovers social network structures from mobile user interaction data and online social network data, generates influencer scores without manual intervention, and applies these scores across marketing and churn management functions autonomously
3Measurement precision
If existing solutions do not combine connectivity structure from diverse sources with mobile user interaction data, then data processing is simpler, but the accuracy and usefulness of influencer scoring is limited
Solution Approach 1:
The patent merges connectivity structures from mobile user interaction data, online social network data, and email network data into a unified social graph model, enabling more accurate influencer scoring by considering multiple data sources simultaneously
Solution Approach 2:
The system creates a composite view of user connections by integrating data from heterogeneous sources (mobile networks, social networks, email networks) to form a comprehensive social graph that captures diverse interaction patterns for more accurate influencer identification
Data Source
AI summary
Embodiments herein provide a method and system that determines an underlying social network from user interaction graphs based on mobile service usage and derive social influence scores for various contexts based on user interaction parameters. The present disclosure pertains to a method of determining social influence score for a user of a social network, said method comprising creating, using a social network analyzer, a social graph based on usage data generated by a plurality of users, said usage data being obtained from a mobile communication network, wherein vertices of the created social graph represent the plurality of users along with edge weights that are based on weighted linear or non-linear combinations of key performance indicators (KPIs) representing actions made by each user; and deriving, by the social network analyzer, influencer score for each user in the social graph based on user interactions from online social networks and mobile interaction patterns.