Vol.:(0123456789) 1 3
Evolving Systems
https://doi.org/10.1007/s12530-019-09274-9
ORIGINAL PAPER
A genetic algorithm for spatiosocial tensor clustering
Exploiting tensorfow potential
Georgios Drakopoulos
1,2
· Foteini Stathopoulou
3
· Andreas Kanavos
4
· Michael Paraskevas
5
· Giannis Tzimas
5
·
Phivos Mylonas
2
· Lazaros Iliadis
6
Received: 22 May 2018 / Accepted: 19 December 2018
© Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract
Tensor clustering is a knowledge management technique which is well known as a major algorithmic and technological
driver behind a broad applications spectrum. The latter ranges from multimodal social media analysis and geolocation pro-
cessing to analytics tailored for large omic data. However, known exact tensor clustering problems when reduced to tensor
factorization are provably NP hard. This is attributed in part to the volume of data contained in a tensor, proportional to the
product of its dimensions, as well as to the increased interdependency between the tensor entries across its dimensions. One
well studied way to circumvent this inherent difculty is to resort to heuristics. This article presents an enhanced version of
a genetic algorithm tailored for community discovery structure in tensors containing spatiosocial data, namely linguistic and
geolocation data. The objective function as well as the chromosome ftness functions by design take into account elements of
linguistic propagation models. The genetic operators of selection, crossover, and mutation as well as the newly added double
mutation operator work directly on the community level. Moreover, various policies for maintaining gene variability across
generations are studied in an extensive simulation powered by Google TensorFlow. As with its predecessor, the proposed
genetic algorithm has been applied to a dataset consisting of a large number of Tweets and their associated geolocations
from the Grand Duchy of Luxembourg, a historically and de facto trilingual country. The results are compared with those
obtained from the original genetic algorithm and their diferences are interpreted.
Keywords Multilingual social networks · Multimodal social networks · Cross cultural communication · Language variation
models · Tensor clustering · Google TensorFlow · Genetic algorithms · Gene variability · Geolocation data · Spatiosocial
data · Humanistic data · Higher order data
Mathematics Subject Classifcation 05C76 · 05C85 · 05D99 · 62H30 · 91C20 · 91C99
CR Subject Classifcation H.2.8 · G.2.2 · G.3 · M.1
* Georgios Drakopoulos
c16drak@ionio.gr
Foteini Stathopoulou
fstathop@uni.lu
Andreas Kanavos
kanavos@ceid.upatras.gr
Michael Paraskevas
mparask@teiwest.gr
Giannis Tzimas
tzimas@teimes.gr
Phivos Mylonas
fmylonas@ionio.gr
Lazaros Iliadis
iliadis@fmenr.duth.gr
1
Cloudminers Inc., Corfu, Greece
2
Department of Informatics, Ionion University, Corfu, Greece
3
University of Luxembourg, Luxembourg City, Luxembourg
4
Hellenic Open University, Patras, Greece
5
Technological and Educational Institution of Western Greece,
Patras, Greece
6
Forest Informatics Lab, Democritus University of Thrace,
Komotini, Greece