Thomas Bouetou Bouetou

Thomas Bouetou Bouetou

Auto-generated profile, based on publicly available information

Thomas Bouetou Bouetou

Department of Computer Engineering, National Advanced School of Engineering, University of Yaound I, Yaound, Cameroon

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
18 January 2025 in International Journal of Geometric Methods in Modern Physics
Verified
Journal Article
Closed access
MFMarius Landry Foka
Marius Landry Foka
SMSalomon Joseph Mbatakou
Salomon Joseph Mbatakou
RPRomain Nimpa Pefoukeu
Romain Nimpa Pefoukeu
MDMichel Bertrand Ngaha Djiadeu
Michel Bertrand Ngaha Djiadeu
TBThomas Bouetou Bouetou
Thomas Bouetou Bouetou

1 January 2025 in Applied Computational Intelligence and Soft Computing
Verified
Journal Article
Open access
GSGiquel Therance Sassa
Giquel Therance Sassa
AKAurelle Tchagna Kouanou
Aurelle Tchagna Kouanou
Jean Louis Fendji Kedieng EbongueJean Louis Fendji Kedieng Ebongue
Jean Louis Fendji Kedieng Ebongue
ANArnauld FOUNTSOP NZEGHA
Arnauld FOUNTSOP NZEGHA
TBThomas Bouetou Bouetou
Thomas Bouetou Bouetou
YEYves Sébastien Emvudu
Yves Sébastien Emvudu
The accurate prediction of subscriber traffic is a critical challenge in telecommunication networks, as it is essential for making informed decisions to maintain consistent service quality. Various artificial intelligence (AI) methods have been explored for this purpose, with recurrent neural networks (RNNs) being particularly popular. However, RNNs often suffer from the gradient vanishing problem, which reduces their effectiveness in processing long sequences of data. Long short–term memory (LSTM) networks partially address this issue by controlling the propagation of learning errors, but they still face challenges related to convergence speed and residual learning errors. To address these limitations, this study proposes a new hybrid RNN architecture that combines features of Jordan and Frasconi networks for improved subscriber traffic prediction. The proposed architecture integrates two sublayers, a sigmoid and a Gaussian function, within a hidden layer that forms a loop. This configuration enables the network to leverage incremental learning through restricted Coulomb energy (RCE) at the local level, while also benefiting from global learning through backpropagation. The architecture was tested on a dataset consisting of cellular traffic data collected from a telecommunication operator in Cameroon. The results demonstrated that the hybrid RNN model outperforms traditional Jordan and Frasconi networks, achieving a root mean square error (RMSE) of 14.18 and a mean absolute error (MAE) of 8.61 using the Adam optimizer with a batch size of 64. In addition, the hybrid model showed superior performance in terms of RMSE, MAE, and convergence speed compared to the existing models. These findings suggest that the proposed model could support telecom operators in proactive congestion management and resource optimization.


10 November 2025 in IEEE Access
Verified
Journal Article
Open access
JKJ. Chiasseup Keudjeu
J. Chiasseup Keudjeu
ONO. Kengni Ngangmo
O. Kengni Ngangmo
SISylvain Iloga
Sylvain Iloga
TBT. Bouetou Bouetou
T. Bouetou Bouetou
Recommendation systems (RSs) are becoming more and more popular in many application areas. Therefore, the three following main categories of RSs have yet been designed: Content-Based, Collaborative Filtering and Hybrid Filtering. Unfortunately, none of the existing RSs take into account the sequential preferences and the sequential profiles of the system users which can, if considered, enable to make finer recommendations. To overcome this limitation, the current work proposes a Hybrid Filtering RS that uses Hidden Markov Models (HMMs) for chronologically learning the features of the recent sequence of movies watched by each user and the features of the recent sequence of users who watched each movie. Meta-data derived from these HMMs are then used for representing each user and each movie as a numeric feature vector whose components are interpretable. The resulting vectors are finally combined for recommending movies to users. Several movie recommendation experiments realized on the open datasets Movielens-100K, Movielens-1M and Movielens-20M exhibited significantly better performances than those of the existing studies regarding the following metrics: precision, recall, F1 score, Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Normalized Discounted Cumulative Gain (NDCG).

19 October 2025 in International Electronic Journal of Geometry
Verified
Journal Article
Closed access
MFMarius Landry Foka
Marius Landry Foka
MDMichel Bertrand Ngaha Djiadeu
Michel Bertrand Ngaha Djiadeu
TBThomas Bouetou Bouetou
Thomas Bouetou Bouetou
The prescribed Ricci curvature problem consists of finding a Riemannian metric $g$ to satisfy the equation $Ric(g) = T$, for some fixed symmetric $(0,2)$-tensor field $T$ on a differential manifold $M$. In this paper, we define Schouten-like metric as a particular solution of a prescribed Ricci curvature problem, and we classify them on five-dimensional nilpotent Lie groups by establishing a link with algebraic Schouten solitons.