No abstract

Abstract-Scantegrity II is an enhancement for existing paper ballot systems. It allows voters to verify election integrityfrom their ballot selections all the way to the final tallyby noting codes and checking for them online. Voters mark Scantegrity II ballots just as with conventional optical scan, but using a special ballot marking pen. Marking a selection with this pen makes legible an otherwise invisible pre-printed confirmation code. Confirmation codes are independent and random for each potential selection on each ballot.To verify that their individual votes are recorded correctly, voters can look up their ballot serial numbers online and verify that their confirmation codes are posted correctly. The confirmation codes do not allow voters to prove how they voted. However, the confirmation codes constitute convincing evidence of error or malfeasance in the event that incorrect codes are posted online. Correctness of the final tally with respect to the published codes is proven by election officials in a manner that can be verified by any interested party. Thus, compromise of either the chain of custody on the ballots or the software systems cannot undetectably affect the election integrity.Scantegrity II has been implemented and tested in small elections, in which ballots were scanned either at the polling place or centrally later. Preparations for its use in a public sector election have commenced.

No abstract

Network attacks have been very prevalent as their rate is growing tremendously. Both organization and individuals are now concerned about their confidentiality, integrity and availability of their critical information which are often impacted by network attacks. To that end, several previous machine learning-based intrusion detection methods have been developed to secure network infrastructure from such attacks. In this paper, an effective anomaly detection framework is proposed utilizing Bayesian Optimization technique to tune the parameters of Support Vector Machine with Gaussian Kernel (SVM-RBF), Random Forest (RF), and k-Nearest Neighbor (k-NN) algorithms. The performance of the considered algorithms is evaluated using the ISCX 2012 dataset. Experimental results show the effectiveness of the proposed framework in term of accuracy rate, precision, low-false alarm rate, and recall.

We propose and implement a cryptographically end-to-end verifiable (E2E) remote voting system for absentee voters and report on its deployment in a binding municipal election in Takoma Park, Maryland. Remotegrity is a hybrid mail/internet extension to the Scantegrity in-person voting system, enabling secure, electronic return of vote-by-mail ballots. It provides voters with the ability to detect unauthorized modifications to their cast ballots made by either malicious client software, or a corrupt election authority-two threats not previously studied in combination. Not only can the voter detect such changes, they can prove it to a third party without giving up ballot secrecy.