A5057395190- Data Mining Algorithms and Applications
- Rough Sets and Fuzzy Logic
- Data Management and Algorithms
- Artificial Intelligence in Games
- Semantic Web and Ontologies
- Data Quality and Management
- Advanced Database Systems and Queries
- Biomedical Text Mining and Ontologies
- Video Analysis and Summarization
- Sports Analytics and Performance
- Multi-Criteria Decision Making
- Olfactory and Sensory Function Studies
- Advanced Chemical Sensor Technologies
- Algorithms and Data Compression
- Complex Network Analysis Techniques
- Gene expression and cancer classification
- Software Engineering Research
- Bayesian Modeling and Causal Inference
- Web Data Mining and Analysis
- Scientific Computing and Data Management
- Digital Games and Media
- Software System Performance and Reliability
- Neural Networks and Applications
- Imbalanced Data Classification Techniques
- Advanced Malware Detection Techniques
Laboratoire d'Informatique en Images et Systèmes d'Information
2013-2023
Université Claude Bernard Lyon 1
2013-2023
Centre National de la Recherche Scientifique
2011-2023
Equipes Traitement de l'Information et Systèmes
2018-2023
Institut National des Sciences Appliquées de Lyon
2013-2021
Delaware Division of Libraries
2019
Lyon College
2014
Institut National des Sciences Appliquées de Rennes
2013
Universidade Federal de Minas Gerais
2012
Institut national de recherche en informatique et en automatique
2010-2011
"Electronic-sport" (E-Sport) is now established as a new entertainment genre. More and more players enjoy streaming their games, which attract even viewers. In fact, in recent social study, casual were found to prefer watching professional gamers rather than playing the game themselves. Within this context, advertising provides significant source of revenue players, casters (displaying other people's games) platforms. For paper, we crawled, during 100 days, most popular among such...
In parallel to the exponential growth of gaming industry, video game live-streaming is rising as a major form online entertainment. Gathering heterogeneous community, popularity this new media led creation web services just for streaming games, such Twitch. TV. paper, we propose model characterize how streamers and spectators behave, based on their possible actions in Twitch and, using it, perform case study Star craft II spectators. analyze large amount data collected TV's chat order better...
An important goal in researching the biology of olfaction is to link perception smells chemistry odorants. In other words, why do some odorants smell like fruits and others flowers? While so-called stimulus-percept issue was resolved field color vision time ago, relationship between psycho-biology odors remains unclear up present day. Although a series investigations have demonstrated that this exists, descriptive explicative aspects proposed models are currently use require greater...
We investigate the problem of mining numerical data with Formal Concept Analysis. The usual way is to use a scaling procedure -transforming attributes into binary ones- leading either loss information or efficiency, in particular w.r.t. volume extracted patterns. By contrast, we propose directlywork on more precise and efficient way. For that, notions closed patterns, generators equivalent classes are revisited context. Moreover, two algorithms proposed tested an evaluation involving...
Recent developments in the frequent pattern mining framework uses additional measures of interest to reduce set discovered patterns. We introduce a rigorous and efficient approach mine statistically significant, unexpected patterns sequences item sets. The proposed methodology is based on null model for multiple testing procedure extract interest. Experiments replays video game demonstrate scalability efficiency method discover strategies.
The success of electronic sports (eSports), where professional gamers participate in competitive leagues and tournaments, brings new challenges for the video game industry. Other than fun, games must be difficult challenging eSports professionals but still easy enjoyable amateurs. In this article, we consider Multi-player Online Battle Arena (MOBA) particularly, "Defense Ancients 2", commonly known simply as DOTA2. context, a challenge is to propose data analysis methods metrics that help...
In this paper, we investigate the problem of mining numerical data in framework Formal Concept Analysis. The usual way is to use a scaling procedure --transforming attributes into binary ones-- leading either loss information or efficiency, particular w.r.t. volume extracted patterns. By contrast, propose directly work on more precise and efficient way, prove it. For that, notions closed patterns, generators equivalent classes are revisited context. Moreover, two original algorithms proposed...
Subgroup discovery (SD) is the task of discovering interpretable patterns in data that stand out w.r.t. some property interest. Discovering accurately discriminate a class from others one most common SD tasks. Standard approaches literature are based on local pattern discovery, which known to provide an overwhelmingly large number redundant patterns. To solve this issue, set mining has been proposed: instead evaluating quality separately, should consider as whole. The goal small diverse and...
This paper shows how to embed a similarity relation between complex descriptions in concept lattices. We formalize by tolerance relation: objects are grouped within same when having similar descriptions, extending the ability of FCA deal with data. propose two different approaches.~A first classical manner defines discretization procedure. A second way consists representing data pattern structures, from which lattice can be constructed directly. In this case, considering mathematically...
The problem addressed in this paper is the merging of numerical information provided by several sources. Merging conflicting pieces into an interpretable and useful format a tricky task even when fusion method chosen. use formal concept analysis pattern structures enables us to associate subsets sources combination results obtainable from consistent information. This provides lattice arguments where reliability can be taken account. Instead providing unique result, yields structured view...
Whereas purest strategic games such as Go and Chess seem timeless, the lifetime of a video game is short, influenced by popular culture, trends, boredom, technological innovations. Even important budget developments allocated editors cannot guarantee timeless success. Instead, novelties corrections are proposed to extend an inevitably bounded lifetime. Novelties can unexpectedly break balance game, players discover unbalanced strategies that developers did not take into account. In new...
To describe the dynamics taking place in networks that structurally change over time, we propose an approach to search for attributes whose value changes impact topology of graph. In several applications, it appears variations a group are often followed by some structural graph one may assume they generate. We formalize triggering pattern discovery problem as method jointly rooted sequence mining and analysis. apply our on three real-world dynamic graphs different natures - co-authoring...
Biclustering numerical data became a popular data-mining task in the beginning of 2000's, especially for analysing gene expression data. A bicluster reflects strong association between subset objects and attributes object/attribute data-table. So called biclusters similar values can be thought as maximal sub-tables with close values. Only few methods address complete, correct non redundant enumeration such patterns, which is well-known intractable problem, while no formal framework exists....