DC Field | Value | Language |
dc.contributor.author | MENAA, Ilyas | - |
dc.contributor.author | BERKANI, AMir SEif El ISlam | - |
dc.date.accessioned | 2024-09-24T09:28:10Z | - |
dc.date.available | 2024-09-24T09:28:10Z | - |
dc.date.issued | 2024 | - |
dc.identifier.uri | https://repository.esi-sba.dz/jspui/handle/123456789/659 | - |
dc.description | Encadrant : Mr MAHAMMED Nadir Co-encadrante : Mme SAIDI Imene | en_US |
dc.description.abstract | The fake news on social media and various other media is widely spreading and is a
matter of serious concern due to its ability to spread false information that can lead to
harmful impacts.
This thesis presents a comprehensive solution for detecting fake news on social media,
utilizing advanced techniques in metaheuristics and machine learning. The proposed
system leverages game based metaheuristic algorithms to optimize machine learning
algorithms which in turn enhance the accuracy and reliability of fake news detection.
Through the use of these state-of-the-art optimization techniques and after thorough
testing using ten datasets, four feature extraction techniques and three novel game-based
metaheuristics, we found that this proposed solution is proved to be effective in increasing
machine learning performance.
Ultimately, the results of this work significantly contributed to combating the spread of
fake news and preventing the potential harm and threat associated with it. ***
Les Informations sur les r´eseaux sociaux et divers autres m´edias se propagent largement
et constituent une pr´eoccupation s´erieuse en raison de leur capacit´e `a diffuser de fausses
informations pouvant entraˆıner des impacts n´efastes.
Ce memoire pr´esente une solution compl`ete pour d´etecter les fausses informations
sur les r´eseaux sociaux, en utilisant des techniques avanc´ees de m´etaheuristiques et
d’apprentissage automatique. Le syst`eme propos´e tire parti des algorithmes m´etaheuristiques
bas´es sur des jeux pour optimiser les algorithmes d’apprentissage automatique, ce
qui am´eliore `a son tour la pr´ecision et la fiabilit´e de la d´etection des fausses nouvelles.
Grˆace `a l’utilisation de ces techniques d’optimisation `a la pointe de la technologie et apr`es
des tests approfondis utilisant dix collections de donn´ees, quatre techniques d’extraction
de caract´eristiques et trois nouvelles m´etaheuristiques bas´ees sur des jeux, nous avons
constat´e que cette solution propos´ee s’est r´ev´el´ee efficace, augmentant les performances de
l’apprentissage automatique.
En fin de compte, les r´esultats de ce travail ont contribu´e de mani`ere significative `a
lutter contre la propagation des fausses nouvelles et `a pr´evenir les dommages et menaces
potentielles associ´es. | en_US |
dc.language.iso | en | en_US |
dc.subject | Fake News Detection | en_US |
dc.subject | Metaheuristics | en_US |
dc.subject | Game-Based Algorithms | en_US |
dc.subject | Machine Learning | en_US |
dc.title | A solution to spot fake news on social media | en_US |
dc.type | Thesis | en_US |
Appears in Collections: | Ingénieur
|