Browse Prior Art Database

System for Handling Location-Puppetry using Social Network Analysis Disclosure Number: IPCOM000238506D
Publication Date: 2014-Aug-31
Document File: 5 page(s) / 169K

Publishing Venue

The Prior Art Database


Disclosed are methods to manage the negative effects of location-puppetry using a modelling system.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 26% of the total text.

Page 01 of 5

System for Handling Location-Puppetry using Social Network Analysis

As the Internet is populated with information and posts from both reputable, professional sources and layperson, unprofessional sources (e.g., posts on social networks). In order to help a user find the most reliable source and information on the topic of interest, systems can use location information. Location information that is tagged with the post is a means of inferring the quality of the post. For example, if User

A is interested in tourist locations in Florida, geolocation information is currently used to help find the best post about that topic. Thus, perceived reputation and trust levels for the post are affected for User A.

People, organizations, or other entities can change an associated reported location in order to affect the ranking/prominence/reputation of the post. This is called location puppetry. For example, a stealth marketer creates one or more pseudonymous accounts, each with modified geolocation information, and then issues posts to further the sales objectives. The different location information changes the post ranking by the search engine and influences the perceived authenticity and trust level for the reader. These puppets publish comments on blogs, wikis, and other public venues about some phenomenon or product in order to generate public interest and influence readers. Location puppetry, especially with harmful intent, represents significant costs to communities and businesses such as revenue loss, tarnished reputation, and marketing impairment.

As more people consume ranking sites for information and recommendations and

consume social media sites for information about events in real-time, there is a need to

determine the true value of these comments postings.

The novel contribution is a system and method to identify location puppetry and then, once a likelihood of location puppetry is determined, assign value weights to a user's posts/comments. This weighting system scores the relative influence of the user's posts/comments.

Existing art does not provide users and site administrators with a likelihood estimate of location puppetry, nor does it post relative relevance. There is no art blocking the specific aspects of this solution.

At a high level, the steps for implementing the system and method are:

1. Determine the likelihood of a user engaged in location puppetry

A. Validating the posts/comments based on location awareness B. Link those with time constraints or as identified by spam classifiers (e.g., Bayesian, Stacking, Cross Validation, Nested Cross validation, etc.)

2. Assign a weighting system to a user's posts/comments

3. Score the relative relevance of the user's posts/comments based on the

weighting system

4. Introduced thresholds for:

A. Likelihood of location puppetry

B. A low score, upon which the system takes remedial action


Page 02 of 5

The system provides a visual cue that allows a user/administrator to infer the e...