Jean-Valère Cossu

NLP-Based Classifiers to Generalize Expert Assessments in E-Reputation

Online Reputation ManagementORM is currently dominated by expert abilities. One of the great challenges is to effectively collect annotated training samples, especially to be able to generalize a small pool of expert feedback from area scale to a more global scale. One possible solution is to use advanced Machine Learning ML techniques, to select annotations from training samples, and propagate effectively and concisely. We focus on the critical issue of understanding the different levels of annotations. Using the framework proposed by the RepLab contest we present a considerable number of experiments in Reputation Monitoring and Author Profiling. The proposed methods rely on a large variety of Natural Language Processing NLP methods exploiting tweet contents and some background contextual information. We show that simple algorithms only considering tweets content are effective against state-of-the-art techniques.

Automatic Classification and PLS-PM Modeling for Profiling Reputation of Corporate Entities on Twitter

In this paper, we address the task of detecting the reputation alert in social media updates, that is, deciding whether a new-coming content has strong and immediate implications for the reputation of a given entity. This content is also submitted to a standard typology of reputation dimensions that consists in a broad classification of the aspects of an under public audience company. Reputation manager needs a real-time database and method to report what is happening right now to his brand. However, typical Natural Language Processing (NLP) approaches to these tasks require external resources and show non-relational modeling. We propose a fast supervised approach for extracting textual features, which we use to train simple statistical reputation classifiers. These classifiers outputs are used in a Partial Least Squares Path Modeling (PLS-PM) system to model the reputation. Experiments on the RepLab 2013 and 2014 collections show that our approaches perform as well as the state-of-the-art more complex methods.

Multi-dimensional Reputation Modeling Using Micro-blog Contents

In this paper, we investigate the issue of modeling corporate entities’ online reputation. We introduce a bayesian latent probabilistic model approach for e-Reputation analysis based on Dimensions (Reputational Concepts) Categorization and Opinion Mining from textual content. Dimensions to analyze e-Reputation are set up by analyst as latent variables. Machine Learning (ML) Natural Language Processing (NLP) approaches are used to label large sets of text passages. For each Dimension, several estimations of the relationship with each text passage are computed as well as Opinion and Priority. The proposed automatic path modeling algorithm explains Opinion or Priority scores based on selected Dimensions. Model Robustness’ is evaluated over RepLab dataset.