Graduate students of Computer Science Department will present their research work. The purpose of this section is to share our knowledge and get to know each other's research interests. Presentations will last for 10 minutes with 5 minutes questions. Presenters should submit both their presentation document and poster in PDF format until October 1st, 2019 by sending an e-mail to: gsacsd@csd.uoc.gr, with the following Subject: [GSC19] Presentation and Poster submission - < Author's Full Name >. Presenters may request for their poster to be printed by the organizing committee of the conference in their submission e-mail. The poster session will take place between 14:00-15:00 and the poster dimensions should be A0 [84.1 x 118.9 cm] size.
We are developing a math-based simulation of the material flow on conveyor belts during the autonomous selection of recyclable items via a robotic arm. Our objective is to accomplish rigorous predictions of the recyclable items’ that will appear next on the belt. Based on the above predictions, optimal selections of the items to be collected are obtained. The latter also depend on a number of factors, such as the cost of each material, cleanliness status, their size, and configuration of the items on the belt, that is distance of the items from each other and distance from the current position of the robot.
Using celebrities to promote brands (also termed as “celebrity endorsement”) is a popular advertising technique. To avoid suboptimal decisions, managers assess consumer perceptions of similarity between the brand and the potential endorser by using costly, survey-based methods. In this study, we combine textual and structural mining methods to facilitate relevant decisions, by extracting information from Twitter data. We propose four metrics, capturing varying levels of brand-related social media activeness. Our methods are validated against survey data, across eight market sectors, Automobiles, Financial services, Technology, Industrial goods & services, Food & beverage, Media & telecommunications, Personal & household goods and Retail. We show that mining data from Twitter accounts elicits perceptions more accurately in industrial/specialized than consumer/mass-market sectors. We propose low-cost, real-time alternatives to survey-based elicitation methods and offer a foundation for future research advances in exploiting textual and structural information from social media as a means to gain richer insights about consumers.
Equinox is a block programming platform suitable for Software Defined Radio applications. It is written in C++11 and targets mainly realtime applications. To meet the strict latency requirements of modern applications, the scheduler of Equinox analyzes the flowgraph of the SDR application and tries to allocate the processing and memory resources required, in a sophisticated and efficient way, utilizing graph partitioning techniques. The project is licensed under the GPLv3 license and the code is available at https://gitlab.com/equinox-sdr/equinox.
Re-sampling based statistical tests are known to be computationally heavy, but reliable when small sample sizes are available. Despite their nice theoretical properties not much effort has been put to make them efficient. In this paper we treat the case of Pearson correlation coefficient and two independent samples t-test. We propose a highly computationally efficient method for calculating permutation based p-values in these two cases. The method is general and can be applied or be adopted to other similar two sample mean or two mean vectors cases.
Security assurance contributes to confidence in the operation and administration procedures of software systems as well as in the security-related properties and functionality of them. Existing assurance methods offer benchmarks and guidelines for designing and implementing such systems as well as a range of assessments for system vulnerabilities and security (e.g., static analysis, static and dynamic testing, penetration testing). Nevertheless, existing assurance methods either do not support continuous security assessment at all, or do not support it adequately. Τhe purpose of the presented research is to develop a continuous security assurance driven approach supporting the dynamic estimation of security and privacy risks for cyber systems and services based on security and privacy assurance models for such systems. The proposed approach combines continuous runtime monitoring with the assessment methods of existing approaches in a hybrid and customisable manner. Based on such hybrid assessments, it aspires to provide a more comprehensive basis for estimating cyber risks and, potentially provide a basis for insuring systems and their owners against it.