no code implementations • EMNLP 2021 • Clément Christophe, Julien Velcin, Jairo Cugliari, Manel Boumghar, Philippe Suignard
Slow emerging topic detection is a task between event detection, where we aggregate behaviors of different words on short period of time, and language evolution, where we monitor their long term evolution.
1 code implementation • 6 Nov 2023 • Irina Proskurina, Guillaume Metzler, Julien Velcin
In this paper, we describe the University of Lyon 2 submission to the Strict-Small track of the BabyLM competition.
no code implementations • 12 Apr 2023 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
On the other hand, a new family of Mixed Membership Stochastic Block Models (MMSBM) allows to model static labeled networks under the assumption of mixed-membership clustering.
1 code implementation • Advances in Intelligent Data Analysis XXI 2023 • Irina Proskurina, Guillaume Metzler, Julien Velcin
Social media platforms have become popular worldwide.
no code implementations • 12 Dec 2022 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
Finally, we develop a use case of the MPDHP on Reddit data.
no code implementations • 12 Dec 2022 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
Building on recent Dirichlet-Point processes literature, we introduce the Houston (Hidden Online User-Topic Network) model, that jointly considers all those features in a non-parametric unsupervised framework.
no code implementations • 20 Sep 2022 • Ian Davidson, Michael Livanos, Antoine Gourru, Peter Walker, Julien Velcin, S. S. Ravi
Explainable AI (XAI) is an important developing area but remains relatively understudied for clustering.
1 code implementation • 16 Sep 2022 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
In this work, we show that these models are all special cases of a single global framework: the Serialized Interacting Mixed membership Stochastic Block Model (SIMSBM).
no code implementations • 16 Sep 2022 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
Most models of information diffusion online rely on the assumption that pieces of information spread independently from each other.
no code implementations • 29 Jan 2022 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
PDHP also alleviates the hypothesis that textual content and temporal dynamics are perfectly correlated.
no code implementations • 5 Nov 2021 • Clément Christophe, Julien Velcin, Jairo Cugliari, Manel Boumghar, Philippe Suignard
Slow emerging topic detection is a task between event detection, where we aggregate behaviors of different words on short period of time, and language evolution, where we monitor their long term evolution.
1 code implementation • 15 Sep 2021 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
Furthermore, the textual content of a document is not always linked to its temporal dynamics.
1 code implementation • 28 Apr 2021 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
We introduce an efficient method to infer both the entities interaction network and its evolution according to the temporal distance separating interacting entities; together, they form the interaction profile.
1 code implementation • 26 Apr 2021 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
This process allows nonparametric estimation of the number of clusters when partitioning datasets.
1 code implementation • 9 Apr 2020 • Gaël Poux-Médard, Julien Velcin, Sabine Loudcher
Here, we propose a new model, the Interactive Mixed Membership Stochastic Block Model (IMMSBM), which investigates the role of interactions between entities (hashtags, words, memes, etc.)
1 code implementation • 7 Apr 2020 • Robin Brochier, Antoine Gourru, Adrien Guille, Julien Velcin
In this direction, document network embedding methods seem to be an ideal choice for building representations of the scientific literature.
no code implementations • 16 Jan 2020 • Antoine Gourru, Adrien Guille, Julien Velcin, Julien Jacques
We present Regularized Linear Embedding (RLE), a novel method that projects a collection of linked documents (e. g. citation network) into a pretrained word embedding space.
1 code implementation • 10 Jan 2020 • Robin Brochier, Adrien Guille, Julien Velcin
We train these word and topic vectors through our general model, Inductive Document Network Embedding (IDNE), by leveraging the connections in the document network.
no code implementations • 11 Sep 2019 • Clément Christophe, Julien Velcin, Jairo Cugliari, Philippe Suignard, Manel Boumghar
Since datasets with annotation for novelty at the document and/or word level are not easily available, we present a simulation framework that allows us to create different textual datasets in which we control the way novelty occurs.
no code implementations • 28 Feb 2019 • Robin Brochier, Adrien Guille, Julien Velcin
In this extended abstract, we present an algorithm that learns a similarity measure between documents from the network topology of a structured corpus.
1 code implementation • 28 Feb 2019 • Robin Brochier, Adrien Guille, Julien Velcin
Even though SGNS better handles non co-occurrence than GloVe, it has a worse time-complexity.
no code implementations • 18 Dec 2018 • Alberto Lumbreras, Julien Velcin, Marie Guégan, Bertrand Jouve
We present a dual-view mixture model to cluster users based on their features and latent behavioral functions.
no code implementations • 14 Jun 2018 • Ciprian-Octavian Truică, Julien Velcin, Alexandru Boicea
In this paper we present a statistical method for automatic language identification of written text using dictionaries containing stop words and diacritics.
no code implementations • JEPTALNRECITAL 2017 • Max Belign{\'e}, Aleks Campar, ra, Jean-Hugues Chauchat, Melanie Lefeuvre, Isabelle Lefort, Sabine Loudcher, Julien Velcin
Cet article s{'}int{\`e}gre dans un projet collaboratif qui vise {\`a} r{\'e}aliser une analyse longitudinale de la production universitaire en G{\'e}ographie.
no code implementations • 11 Jan 2016 • Marian-Andrei Rizoiu, Julien Velcin, Stéphane Lallich
In this paper, we propose a new time-aware dissimilarity measure that takes into account the temporal dimension.
no code implementations • 11 Jan 2016 • Young-Min Kim, Julien Velcin, Stéphane Bonnevay, Marian-Andrei Rizoiu
Evolutionary clustering aims at capturing the temporal evolution of clusters.
no code implementations • 17 Dec 2015 • Marian-Andrei Rizoiu, Julien Velcin, Stéphane Lallich
We seek to construct, in an unsupervised way, new features that are more appropriate for describing a given dataset and, at the same time, comprehensible for a human user.
no code implementations • 14 Dec 2015 • Marian-Andrei Rizoiu, Julien Velcin, Stéphane Lallich
We apply our proposition to the task of content-based image classification and we show that semantically enriching the image representation yields higher classification performances than the baseline representation.
no code implementations • 24 Sep 2015 • Md. Abul Hasnat, Julien Velcin, Stéphane Bonnevay, Julien Jacques
In this paper, we propose a novel evolutionary clustering method based on the parametric link among Multinomial mixture models.
no code implementations • JEPTALNRECITAL 2015 • Leila Khouas, Caroline Brun, Anne Peradotto, Jean-Val{\`e}re Cossu, Julien Boyadjian, Julien Velcin
Ce travail concerne l{'}int{\'e}gration {\`a} une plateforme de veille sur internet d{'}outils permettant l{'}analyse des opinions {\'e}mises par les internautes {\`a} propos d{'}une entit{\'e}, ainsi que la mani{\`e}re dont elles {\'e}voluent dans le temps.
no code implementations • 9 May 2015 • Md. Abul Hasnat, Julien Velcin, Stéphane Bonnevay, Julien Jacques
In this paper, we study different discrete data clustering methods, which use the Model-Based Clustering (MBC) framework with the Multinomial distribution.
2 code implementations • 28 Apr 2015 • Marian-Andrei Rizoiu, Adrien Guille, Julien Velcin
Constructed as a web platform, CommentWatcher features automatic mass fetching of user posts from forum on multiple sites, extracting topics, visualizing the topics as an expression cloud and exploring their temporal evolution.
no code implementations • LREC 2014 • Julien Velcin, Young-Min Kim, Caroline Brun, Jean-Yves Dormagen, Eric SanJuan, Leila Khouas, Anne Peradotto, Stephane Bonnevay, Claude Roux, Julien Boyadjian, Alej Molina, ro, Marie Neihouser
The objective of this paper is to describe the design of a dataset that deals with the image (i. e., representation, web reputation) of various entities populating the Internet: politicians, celebrities, companies, brands etc.