Search Results for author: Julien Velcin

Found 27 papers, 9 papers with code

Monitoring geometrical properties of word embeddings for detecting the emergence of new topics.

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.

Event Detection Word Embeddings

Le Processus Powered Dirichlet-Hawkes comme A Priori Flexible pour Clustering Temporel de Textes

no code implementations29 Jan 2022 Gaël Poux-Médard, Julien Velcin, Sabine Loudcher

PDHP also alleviates the hypothesis that textual content and temporal dynamics are perfectly correlated.

Monitoring geometrical properties of word embeddings for detecting the emergence of new topics

no code implementations5 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.

Event Detection Word Embeddings

Powered Hawkes-Dirichlet Process: Challenging Textual Clustering using a Flexible Temporal Prior

1 code implementation15 Sep 2021 Gaël Poux-Médard, Julien Velcin, Sabine Loudcher

Furthermore, the textual content of a document is not always linked to its temporal dynamics.

Information Interaction Profile of Choice Adoption

1 code implementation28 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.

Interactions in information spread: quantification and interpretation using stochastic block models

1 code implementation9 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.)

Stochastic Block Model

New Datasets and a Benchmark of Document Network Embedding Methods for Scientific Expert Finding

1 code implementation7 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.

Network Embedding

Document Network Projection in Pretrained Word Embedding Space

no code implementations16 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.

General Classification Information Retrieval +3

Inductive Document Network Embedding with Topic-Word Attention

1 code implementation10 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.

Network Embedding

How to detect novelty in textual data streams? A comparative study of existing methods

no code implementations11 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.

Global Vectors for Node Representations

1 code implementation28 Feb 2019 Robin Brochier, Adrien Guille, Julien Velcin

Even though SGNS better handles non co-occurrence than GloVe, it has a worse time-complexity.

Network Embedding

Link Prediction with Mutual Attention for Text-Attributed Networks

no code implementations28 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.

Link Prediction

Non-parametric clustering over user features and latent behavioral functions with dual-view mixture models

no code implementations18 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.

Automatic Language Identification for Romance Languages using Stop Words and Diacritics

no code implementations14 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.

Language Identification

How to Use Temporal-Driven Constrained Clustering to Detect Typical Evolutions

no code implementations11 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.

Unsupervised Feature Construction for Improving Data Representation and Semantics

no code implementations17 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.

Two-sample testing

Semantic-enriched Visual Vocabulary Construction in a Weakly Supervised Context

no code implementations14 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.

General Classification Image Classification

Opinion mining from twitter data using evolutionary multinomial mixture models

no code implementations24 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.

Opinion Mining

Etude de l'image de marque d'entit\'es dans le cadre d'une plateforme de veille sur le Web social

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.

Simultaneous Clustering and Model Selection for Multinomial Distribution: A Comparative Study

no code implementations9 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.

Model Selection

CommentWatcher: An Open Source Web-based platform for analyzing discussions on web forums

2 code implementations28 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.

Investigating the Image of Entities in Social Media: Dataset Design and First Results

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.

Information Retrieval Opinion Mining +1

Cannot find the paper you are looking for? You can Submit a new open access paper.