Nicolas Dugué

Nicolas Dugué


Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2015
Verified
Conference Paper
Open access
JCJean-Valere Cossu
Jean-Valere Cossu
NDNicolas Dugué
Nicolas Dugué
VLVincent Labatut
Vincent Labatut
In this paper, we investigate the issue of detecting the real-life influence of people based on their Twitter account. We propose an overview of common Twitter features used to characterize such accounts and their activity, and show that these are inefficient in this context. In particular, retweets and followers numbers, and Klout score are not relevant to our analysis. We thus propose several Machine Learning approaches based on Natural Language Processing and Social Network Analysis to label Twitter users as Influencers or not. We also rank them according to a predicted influence level. Our proposals are evaluated over the CLEF RepLab 2014 dataset, and outmatch state-of-the-art ranking methods.

21 June 2022 in Physica A: Statistical Mechanics and its Applications
Verified
Journal Article
Closed access
NDNicolas Dugué
Nicolas Dugué
APAnthony Perez
Anthony Perez

9 May 2016 in Social Network Analysis and Mining
Verified
Journal Article
Open access
PDF file
JCJean-Valère Cossu
Jean-Valère Cossu
VLVincent Labatut
Vincent Labatut
NDNicolas Dugué
Nicolas Dugué
Many works related to Twitter aim at characterizing its users in some way: role on the service (spammers, bots, organizations, etc.), nature of the user (socio-professional category, age, etc.), topics of interest, and others. However, for a given user classification problem, it is very difficult to select a set of appropriate features, because the many features described in the literature are very heterogeneous, with name overlaps and collisions, and numerous very close variants. In this article, we review a wide range of such features. In order to present a clear state-of-the-art description, we unify their names, definitions and relationships, and we propose a new, neutral, typology. We then illustrate the interest of our review by applying a selection of these features to the offline influence detection problem. This task consists in identifying users who are influential in real life, based on their Twitter account and related data. We show that most features deemed efficient to predict online influence, such as the numbers of retweets and followers, are not relevant to this problem. However, we propose several content-based approaches to label Twitter users as influencers or not. We also rank them according to a predicted influence level. Our proposals are evaluated over the CLEF RepLab 2014 dataset, and outmatch state-of-the-art methods.

11 March 2014 in Social Network Analysis and Mining
Verified
Journal Article
Closed access
PDF file
NDNicolas Dugué
Nicolas Dugué
APAnthony Perez
Anthony Perez

3 November 2016
Verified
Conference Paper
Open access
JLJean-Charles Lamirel
Jean-Charles Lamirel
NDNicolas Dugué
Nicolas Dugué
PCPascal Cuxac
Pascal Cuxac
This paper deals with a major challenge in clustering that is optimal model selection. It presents new efficient clustering quality indexes relying on feature maximization, which is an alternative measure to usual distributional measures relying on entropy, Chi-square metric or vector-based measures such as Euclidean distance or correlation distance. First Experiments compare the behavior of these new indexes with usual cluster quality indexes based on Euclidean distance on different kinds of test datasets for which ground truth is available. This comparison clearly highlights altogether the superior accuracy and stability of the new method on these datasets, its efficiency from low to high dimensional range and its tolerance to noise. Further experiments are then conducted on “real life” textual data extracted from a multisource bibliographic database for which ground truth is unknown. These experiments show that the accuracy and stability of these new indexes allow to deal efficiently with diachronic analysis, when other indexes do not fit the requirements for this task.

14 June 2015 in Social Network Analysis and Mining
Verified
Journal Article
Open access
PDF file
NDNicolas Dugué
Nicolas Dugué
VLVincent Labatut
Vincent Labatut
APAnthony Perez
Anthony Perez
In the context of Twitter, social capitalists are specific users trying to increase their number of followers and interactions by any means. These users are not healthy for the service, because they are either spammers or real users flawing the notions of influence and visibility. Studying their behavior and understanding their position in Twitter is thus of important interest. It is also necessary to analyze how these methods effectively affect user visibility. Based on a recently proposed method allowing to identify social capitalists, we tackle both points by studying how they are organized, and how their links spread across the Twitter follower–followee network. To that aim, we consider their position in the network w.r.t. its community structure. We use the concept of community role of a node, which describes its position in a network depending on its connectivity at the community level. However, the topological measures originally defined to characterize these roles consider only certain aspects of the community-related connectivity, and rely on a set of empirically fixed thresholds. We first show the limitations of these measures, before extending and generalizing them. Moreover, we use an unsupervised approach to identify the roles, in order to provide more flexibility relatively to the studied system. We then apply our method to the case of social capitalists and show they are highly visible on Twitter, due to the specific roles they hold.