In modern times it seems plausible to use social media to influence the results of an election. This possibility represents a threat to any democracy and our aim is to understand when this is actually feasible so as to understand when it is necessary to find appropriate countermeasures. This problem has been recently treated in the literature, but main works on it assume a complete knowledge on the preference profiles of the voters involved in the election. Instead, in this thesis a step towards modeling uncertainty in this field is made. In fact, a plausible solution to the problem of election manipulation with unknown preference profiles is been proposed and analysed. Many experimental results are reported with several different settings. These outcomes make clear that different manipulation approaches have to be applied for solving different scenario (e.g. different network structure).
Nei tempi moderni sembra plausibile usare i social media per influenzare i risultati di un’elezione. Questa possibilità rappresenta una minaccia per qualsiasi democrazia e il nostro obiettivo è quello di capire quando questo è effettivamente possibile per capire quando è necessario trovare contromisure adeguate. Questo problema è stato recentemente trattato in letteratura, ma i principali lavori su di esso ipotizzano una conoscenza completa delle preferenze degli elettori chiamati a votare. Invece, in questa tesi si fa un passo verso la modellazione dell’incertezza in questo campo. Infatti, viene proposta e analizzata una soluzione plausibile al problema della manipolazione delle elezioni con preferenze sconosciute. Molti risultati sperimentali sono riportati con diverse impostazioni. Questi risultati rendono chiaro che diversi approcci di manipolazione devono essere applicati per risolvere diversi scenari (e.g. diversa struttura della rete).
Election manipulation with unknown voters’ preferences
Vitellio, Carlo Augusto
2020/2021
Abstract
In modern times it seems plausible to use social media to influence the results of an election. This possibility represents a threat to any democracy and our aim is to understand when this is actually feasible so as to understand when it is necessary to find appropriate countermeasures. This problem has been recently treated in the literature, but main works on it assume a complete knowledge on the preference profiles of the voters involved in the election. Instead, in this thesis a step towards modeling uncertainty in this field is made. In fact, a plausible solution to the problem of election manipulation with unknown preference profiles is been proposed and analysed. Many experimental results are reported with several different settings. These outcomes make clear that different manipulation approaches have to be applied for solving different scenario (e.g. different network structure).File | Dimensione | Formato | |
---|---|---|---|
2021_12_Vitellio.pdf
solo utenti autorizzati dal 10/12/2024
Dimensione
1.44 MB
Formato
Adobe PDF
|
1.44 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/181753