top of page
  • Writer's pictureOssi Mokryn

Domain-based Latent Personal Analysis (LPA) and its uses

Updated: Jun 8, 2021

LPA is an easy-to-use and fast domain-based spectral signature that can be used in a variety of domains that is tailored for big-data computations. It creates a strong and unique signature of both over-used and missing items that identify a user or a component in a domain. It can be used in a variety of domains, from textual to computational.


Paper: Mokryn, O., & Ben-Shoshan, H. (2020). Domain-based Latent Personal Analysis and its use for impersonation detection in social media. Accepted to User Modeling and User Adapt Interaction (UMUAI) May 2021. ArXiv preprint arXiv:2004.02346. (PDF)


Domain-based Latent Personal Analysis and its use for impersonation detection in social media Zipf's law defines an inverse proportion between a word's ranking in a given corpus and its frequency in it, roughly dividing the vocabulary to frequent (popular) words and infrequent ones. Here, we stipulate that within a domain an author's signature can be derived from, in loose terms, the author's missing popular words and frequently used infrequent words. We devise a method, termed Latent Personal Analysis (LPA), for finding such domain-based personal signatures. LPA  determines what words most contributed to the distance between a user's vocabulary from the domain's. We identify the most suitable distance metric for the method among several and construct a personal signature for authors. We validate the correctness and power of the signatures in identifying authors and utilize  LPA to identify two types of impersonation in social media: (1) authors with sockpuppets (multiple) accounts; (2)  front-user accounts, operated by several authors.  We validate the algorithms and employ them over a large scale dataset obtained from a social media site with over 4000 accounts, and corroborate the results employing temporal rate analysis. LPA can be used to devise personal signatures in a wide range of scientific domains in which the constituents have a long-tail distribution of elements.


LPA for music: Spotify 2017 streaming


Authorship attribution applications:




Impersonation on social media - Sockpuppets accounts

LPA is fast and easy to implement at a large-scale. We then deployed it over 4000 IMDb reviewer accounts to find sockpuppets accounts activated by a single author. We then performed a 4000x4000 similarity measured between all accounts' LPA signatures.

The full list of suspected Sockpuppets

Example 1

User 59775972 (joshuadrake-39480) and User 62431316 (joshuadrake-91275) have 502 common terms in their LPA-signatures and the distance between their signatures is 0.013.


Example 2

User 24051675 (jpachar82) and User 53564354 (jasonpachar) have 584 common terms in their LPA-signatures and the distance between their signatures is 0.2965892






 

157 views3 comments

3 Comments


Steele Nickle
Steele Nickle
Sep 10

You have to unlock the homes in order to meet the following geometry dash subzero characters.


Like

dromond melody
dromond melody
Aug 26

What are the emotional responses elicited during the consumption of experiential commodities, such as films, songs, or restaurants? geometry dash online

Like

Poppy Mis
Poppy Mis
Aug 24

Hours of delicious fun await those who are both seasoned puzzle enthusiasts and casual gamers seeking light entertainment with the quordle game.

Like
bottom of page