Title :

Evolutionary clustering for categorical data using parametric links among multinomial mixture models

Author(s) :

Hasnat, Md Abul [Auteur]
Equipe de Recherche en Ingénierie des Connaissances [ERIC]
Velcin, Julien [Auteur]
Equipe de Recherche en Ingénierie des Connaissances [ERIC]
Entrepôts, Représentation et Ingénierie des Connaissances [ERIC]
Bonnevay, Stephane [Auteur]
Equipe de Recherche en Ingénierie des Connaissances [ERIC]
Jacques, Julien [Auteur]
MOdel for Data Analysis and Learning [MODAL]
Equipe de Recherche en Ingénierie des Connaissances [ERIC]
Laboratoire Paul Painlevé - UMR 8524 [LPP]
Entrepôts, Représentation et Ingénierie des Connaissances [ERIC]

Journal title :

Econometrics and Statistics

Pages :

141-159

Publisher :

Elsevier

Publication date :

2017-07

ISSN :

2452-3062

English keyword(s) :

Twitter data
evolutionary clustering
multinomial distribution
model-based clustering
mixture model

HAL domain(s) :

Mathématiques [math]/Statistiques [math.ST]
Informatique [cs]/Web
Informatique [cs]/Traitement du texte et du document
Informatique [cs]/Intelligence artificielle [cs.AI]
Informatique [cs]/Recherche d'information [cs.IR]

English abstract : [en]

In this paper, we propose a novel evolutionary clustering method for temporal categorical data based on parametric links among multinomial mixture models. Besides clustering, our main goal is to interpret the evolutions ...
Show more >In this paper, we propose a novel evolutionary clustering method for temporal categorical data based on parametric links among multinomial mixture models. Besides clustering, our main goal is to interpret the evolutions of clusters over time. To this aim, first we propose the formulation of a generalized model that establishes parametric links among two multinomial mixture. Afterward, different parametric sub-models are defined in order to model typical evolutions of the clustering structure. Model selection criteria allow to select the best sub-models and thus to guess the clustering evolution.For the experiments, first we evaluate the proposed method with synthetic temporal data. Next, we apply it to analyze the annotated social media data. Results show that the proposed method is better than the state-of-the-art based on the common evaluation metrics. Additionally, it can provide interpretation about the temporal evolution of the clusters.Show less >

Language :

Anglais

Popular science :

Non

ANR Project :

Images sur le Web : analyse de la dynamique des images sur le Web 2.0.

Collections :

• Laboratoire Paul Painlevé - UMR 8524

Source :

Harvested from HAL