Christophe Thovex

Semantic Social Networks Analysis : Towards a Sociophysical Knowledge Analysis

In 1977, Freeman formalised generic measures of social networks analysis (SNA). Then, the Web 2.0 social networks have become global networks (e.g., FaceBook, MSN). This article presents a semantic model, non probabilistic and predictive, for the decisional analysis of professional and institutional social networks. The presented multidisciplinary model, in parallel to Galam sociophysics, integrates some semantic methods of knowledge engineering and natural language processing, some measures of statistic sociology and some electrodynamic laws, applied to the economic performance and social climate optimisation. It is currently under experimentation, in line with the Socioprise project, funded by the French State Secretariat at the prospective and development of the digital economy.

Static, dynamic and semantic dimensions: towards a multidisciplinary approach of social networks analysis

The objective of our work is to extend static and dynamic models of Social Networks Analysis (SNA), by taking conceptual aspects of enterprises and institutions social graph into account. The originality of our multidisciplinary work is to introduce abstract notions of electrophysic to define new measures in SNA, for new decision-making functions dedicated to Human Resource Management (HRM). This paper introduces a multidimensional system and new measures: (1) a tension measure for social network analysis, (2) an electrodynamic, predictive and semantic system for recommendations on social graphs evolutions and (3) a reactance measure used to evaluate the individual stress at work of the members of a social network.

An Epistemic Equivalence for Predictive Social Networks Analysis

The paper presents a semantic and dynamic model providing predictive recommendations and trend predictions, based on the analysis of social networks and social media. It introduces a significant completion of the interdisciplinary paradigm based on the analogy of information flows and current, commonly used in social networks analysis. It defines a social graph structure entailing knowledge engineering, which contributes to semantic clustering and classification of the Social Web and participates in a model of predictive recommendations based on natural laws in electro-physics. These multidisciplinary contributions are applied to team performance and social climate optimization, within collaborative organizations (patented model). Outcomes are presented in line with the Socioprise project, funded by the French State Secretariat at the prospective and development of the digital economy.

Opinion mining and semantic analysis of touristic social networks

In the context of a Regional Platform of Innovation for future tourism, we define a model and a decision support system for semantic analysis of social networks embedding networks of opinions, aiming at representing and understanding territorial uses coming out from digital marks. The territorial and touristic observatory we develop behind a digital platform is designed to complete usual tools of economic intelligence, helping in governance and in the identification of touristic products and services of tomorrow, so as to foster the rise of territorial economy. In the early stage of the project, the contributions we present are (1) a social semantic graph structure, (2) the definition of semantic degree centrality, (3) a process and definitions for interfacing semantic networks of opinions with social semantic networks of uses, and (4) a first use case experimentation. Public deployment is planned for 2013 and economical impacts will be measured on next years.

A Semantic and Multidisciplinary Model for Professional and Social Networks Analysis

By bridge-building between the classical models of social networks analysis, ontologies engineering and physics, our work defines a multidisciplinary model of professional social networks analysis, dedicated to human and social capital management in enterprises and institutions. We introduce a semantic process of social graphs static and dynamic analysis, based on the enterprise content and producing decisional tools for the performance management. This approach is in line with the Socio prise project, in partnership with a leader software engineering company for human capital management. Socio prise is funded by the French State Secretariat at the prospective and development of digital economy.

A Multidisciplinary Model of Dynamic and Semantic Social Networks Analysis for Institutions

Social networks of the Web 2.0 have become global (e.g. FaceBook, MSN). In 1977, L. C. FREEMAN published the first generic metrics for Social Networks Analysis (SNA), mainly based on static graph-mining models. The objective of our work is to introduce new dynamic SNA models dedicated to SNA and to take the conceptual aspects of enterprises and institutions social graph into account. Our work is based on the definition of new multidimensional measures in SNA for new decision-making functions in Human Resource Management (HRM). The presented contributions makes it original. They are: (1) a measure of tension of a social network, (2) an electrodynamic and predictive system for semantic recommendations on social graphs evolutions and (3) a measure of reactance of a social network used to evaluate the individual stress of its members.

Automatic Building of Socio-semantic Networks for Requirements Analysis

Experts teams using large volumes of official documents need to see and understand at a glance how texts regarding a topic are redundant and depend on each other. In line with the strategic line of a consultants company, we present a decision support system for the visual analysis of requirements and regulation texts, based on a new model of semantic social networks analysis. We present our model and a business application. In this work, standard metrics of semantic and linguistic statistics are combined with measures of social networks analysis, in order to display socio-semantic networks of regulation texts supporting experts’ decision.