My academic profiles

List of publications

  1. Diniz Augusto, R.A., Ghassany, M., Chaieb, F., and Hadj Selem, F. 2024. Active Learning with Unfiltered Informativeness Technique for Object Detection. IFIP International Conference on Artificial Intelligence Applications and Innovations, Springer, 252–262.

    Contemporary Deep Learning models demand substantial volumes of data to effectively learn, creating a challenge given the difficulty of obtaining well-annotated data. Moreover, not all samples within a dataset are of equal significance to the learning process. Active Learning emerges as a solution to this issue by providing a structured framework for selecting the most instructive data within a dataset. This approach involves isolating a subset of the N most informative samples to train the algorithm effectively. In response to this challenge, we introduce Unfiltered Informativeness, a novel framework designed to assess the informativeness value of samples within a dataset. Our approach employs a trained detector to identify objects in a scene and subsequently computes the information gain of these objects. When multiple objects are present within a scene, we aggregate multiple scores into a single informative score. We systematically evaluate our approach against Random Sampling and other strategies.

  2. Brue, G., Chaieb, F., Dantan, J., et al. 2024. Decision-Making Approach for Early Plant Stress Detection from Hyperspectral Images. ACIIDS 2024, Springer Nature Singapore, 181–192.

    Smart agriculture is based on advanced technologies such as artificial intelligence to enhance agricultural efficiency while minimizing the use of resources. Collecting multimodal data provides inputs for both predictive models and decision support. In this context, hyperspectral (HS) images aim to support crop monitoring, facilitate disease detection processes, and tackle water scarcity challenges by promptly identifying water stress. The domain of computer vision-assisted smart agriculture is continually progressing, marked by a substantial volume of recent scientific literature detailing advancements in this domain. This progression aligns with the numerous advancements achieved in the realm of deep learning over the past few years. The objective of this paper is to introduce a methodology for constructing a model that classifies the level of water stress using a dataset comprised of image series of water-stressed plants, eliminating the necessity for additional precise labeling. Once constructed, this model requires only a single hyperspectral image of a plant to ascertain the degree of water stress.

  3. Sabry, S., Ghassany, M., and Anzalone, S.M. 2020. People identity learning in HRI through an incremental multimodal approach. CognitIve RobotiCs for intEraction, CIRCE workshop at the 29th IEEE International Conference on Robot & Human Interactive Communication.

    In order to achieve a real partnership with humans, robots should be able to personalize and adapt their behaviors to the specifity of each human partner. A fundamental step to achieve this goal is to develop a reliable characterisation of their identity. In this paper, we present a robotic system capable of learning the identities of humans through an incremental multimodal approach. Features from faces and voices are extracted as people signature and exploited to achieve a self-supervised learning. A feasibility study helped us to underline challenges, opportunities and possible obstacles in the development of such skills.

  4. Benamar, L., Balagué, C., and Ghassany, M. 2017. The Identification and Influence of Social Roles in a Social Media Product Community. Journal of Computer‐Mediated Communication.

    This research focuses on the identification of social roles and an investigation of their influence in online context. Relying on a systemic approach for role conceptualization, we investigate member’s activity, shared content and position in the network within a consumer to consumer social media-based community (SMC) around a product. This investigation led to the identification of ten core roles, based on three key elements: object of interest (product, practice, and community), main contribution type (sharing information and seeking information), individual orientation (factual, emotional). We propose an explanation about how these roles, through their positioning, participate in the community dynamics and how they contribute to the creation and diffusion of cookery as a social practice, shaping the periphery around this practice.

  5. Benamar, L., Ghassany, M., and Balagué, C. 2016. Analyzing interactions and identifying social roles in a brand community on social networks. The European Marketing Academy, EMAC2016.
  6. Ghassany, M. and Bennani, Y. 2015. Collaborative Fuzzy Clustering of Variational Bayesian Generative Topographic Mapping. International Journal of Computational Intelligence and Applications 14.

    In this paper, we propose a Collaborative Clustering method based on Variational Bayesian Generative Topographic Mapping (VBGTM). To do so, we first propose a method that combines VBGTM and Fuzzy c-means (FCM). Collaborative clustering is useful to achieve interaction between different sources of information for the purpose of revealing underlying structures and regularities within data sets. It can be treated as a process of consensus building where we attempt to reveal a structure that is common across all sets of data. VBGTM was introduced as a variational approximation of Generative Topographic Mapping (GTM) to control data overfitting. It provides an analytical approximation to the posterior probability of the latent variables and the distribution of the input data in the latent space. It can be effectively applied to visualize and explore properties of the data. But when the number of latent points is large, similar units need to be grouped (i.e., clustered) to facilitate quantitative analysis of the map and the data. We use FCM to determine the prototypes as well as the resultant clusters and the corresponding membership functions of the input data, based on the latent variables obtained from VBGTM. So, by combining the two algorithms, we develop a method that can do visualization and clustering at the same time. We observe that the hybrid method (F-VBGTM) performs very well in terms of many cluster-validity indexes.

  7. Ghassany, M., Grozavu, N., and Bennani, Y. 2013. Collaborative Multi-View Clustering. The 2013 International Joint Conference on Neural Networks, IJCNN13, Dallas, USA.
  8. Ghassany, M., Grozavu, N., and Bennani, Y. 2012. Collaborative Clustering Using Prototype-Based Techniques. International Journal of Computational Intelligence and Applications 11.

    The aim of collaborative clustering is to reveal the common structure of data distributed on different sites. In this paper, we present a formalism of topological collaborative clustering using prototype-based clustering techniques; in particular we formulate our approach using Kohonen’s Self-Organizing Maps. Maps representing different sites could collaborate without recourse to the original data, preserving their privacy. We present two different approaches of collaborative clustering: horizontal and vertical. The strength of collaboration (confidence exchange) between each pair of datasets is determined by a parameter, we call coefficient of collaboration, to be estimated iteratively during the collaboration phase using a gradient-based optimization, for both the approaches. The proposed approaches have been validated on several datasets and experimental results have shown very promising performance.

  9. Ghassany, M., Grozavu, N., and Bennani, Y. 2012. Collaborative Generative Topographic Mapping. Neural Information Processing: 19th International Conference, ICONIP 2012, Doha, Qatar.
  10. Grozavu, N., Ghassany, M., and Bennani, Y. 2011. Apprentissage de la confiance des échanges en classification collaborative non supervisée. 7e Plateforme AFIA, Association Française pour l’Intelligence Artificielle, Chambéry, 16 au 20 mai 2011, Editions Publibook, 217.
  11. Grozavu, N., Ghassany, M., and Bennani, Y. 2011. Learning confidence exchange in Collaborative Clustering. The 2011 International Joint Conference on Neural Networks, IJCNN11, San Jose, CA, USA.

PhD

  1. Ghassany, M. 2013. Contributions to Collaborative Clustering. .

Master

  1. Ghassany, M. 2010. The adaptive fused lasso regression and its application on microarrays CGH data. .