Pinned Repositories
Architectural-TD-Metrics-Calculator
Modularity is one of the key principles of software design. In order for a software system to be modular, it should be organized into modules that are highly coherent internally, whereas at the same time as independent from other modules as possible. In this paper we explore coupling and cohesion metrics at the software package level—i.e., one of most basic levels of software functional decomposition in object-oriented (OO) systems, with the aim of investigating their relation to the technical debt of each package. Current state-of-the-art tools in TD measurement are working on the source code level, and the extent to which they can unveil limitations at the architecture level (e.g., violations of the modularity principle), has not been explored so far. To achieve this goal, we conducted a case study on 1,200 packages retrieved from 20 well-known open source software projects. The results of the study suggested that current measures of technical debt are able to identify / predict modules that lack modularity, and therefore suffer from Architectural Technical Debt (ATD). The results of the study are discussed both from the practitioners’ and researchers’ point of view.
Continuous-implicit-authentication-of-smartphone-users-using-navigation-data
This project, proposes a methodology for continuous implicit authentication of smartphones users, using the navigation data, in order to improve the security and ensure the privacy of sensitive personal data.
Convolution-Neural-Network--Image-classification
# Construction of a Convolution Neural Network (CNN) for image classification. The network is going to classify objects from the images from 10 different classes. The dataset used is a subset of CIFAR-100.
Hackathon-Sana-Bootcamp
The goal was to create within 10 -15 hours a smart mobile or tablet application that will provide health solutions. My team and I worked in the expansion of an android application called vida24 from Vidavo. This application includes many health solutions, but our main focus was on the pacemaker.
Move-Class-Refactoring-Tool
The purpose of this diploma thesis was to develop a methodology for quantifying cohesion and coupling at package level in order to propose solutions to improve the design quality of object-oriented systems
PreECESCON-Smart-House
PreECESCON is a hackathon event, part of the Student’s Conference of Electrical and Computer Engineering (ECESCON) that take place in various cities of Greece, since 2007. This year's concept was smart houses, meaning the various ways to make our daily lives more comfortable. We created a simulation of a smart home system with the main purpose of energy saving. More specifically, our project contains sensors in order to detect humidity and temperature, provides user with an easy to use website where he/she can check and change the state of the devices (e.g. close the tv or open the window) and keep him updated about what happens in the house (which devices are on). We came in sixth.
Technical-Debt-Management-Toolbox
Technical Debt Management is a ToolBox developed in the context of a European Union project H2020 (SDK4ED - https://sdk4ed.eu) from University of Macedonia (UoM). It provides analysis in all aspects of technical debt (TD principal, TD interest, TD breaking point, TD interest probability).
AngelikiTsintzira's Repositories
AngelikiTsintzira/Continuous-implicit-authentication-of-smartphone-users-using-navigation-data
This project, proposes a methodology for continuous implicit authentication of smartphones users, using the navigation data, in order to improve the security and ensure the privacy of sensitive personal data.
AngelikiTsintzira/Architectural-TD-Metrics-Calculator
Modularity is one of the key principles of software design. In order for a software system to be modular, it should be organized into modules that are highly coherent internally, whereas at the same time as independent from other modules as possible. In this paper we explore coupling and cohesion metrics at the software package level—i.e., one of most basic levels of software functional decomposition in object-oriented (OO) systems, with the aim of investigating their relation to the technical debt of each package. Current state-of-the-art tools in TD measurement are working on the source code level, and the extent to which they can unveil limitations at the architecture level (e.g., violations of the modularity principle), has not been explored so far. To achieve this goal, we conducted a case study on 1,200 packages retrieved from 20 well-known open source software projects. The results of the study suggested that current measures of technical debt are able to identify / predict modules that lack modularity, and therefore suffer from Architectural Technical Debt (ATD). The results of the study are discussed both from the practitioners’ and researchers’ point of view.
AngelikiTsintzira/PreECESCON-Smart-House
PreECESCON is a hackathon event, part of the Student’s Conference of Electrical and Computer Engineering (ECESCON) that take place in various cities of Greece, since 2007. This year's concept was smart houses, meaning the various ways to make our daily lives more comfortable. We created a simulation of a smart home system with the main purpose of energy saving. More specifically, our project contains sensors in order to detect humidity and temperature, provides user with an easy to use website where he/she can check and change the state of the devices (e.g. close the tv or open the window) and keep him updated about what happens in the house (which devices are on). We came in sixth.
AngelikiTsintzira/Hackathon-Sana-Bootcamp
The goal was to create within 10 -15 hours a smart mobile or tablet application that will provide health solutions. My team and I worked in the expansion of an android application called vida24 from Vidavo. This application includes many health solutions, but our main focus was on the pacemaker.
AngelikiTsintzira/Move-Class-Refactoring-Tool
The purpose of this diploma thesis was to develop a methodology for quantifying cohesion and coupling at package level in order to propose solutions to improve the design quality of object-oriented systems
AngelikiTsintzira/Convolution-Neural-Network--Image-classification
# Construction of a Convolution Neural Network (CNN) for image classification. The network is going to classify objects from the images from 10 different classes. The dataset used is a subset of CIFAR-100.
AngelikiTsintzira/Technical-Debt-Management-Toolbox
Technical Debt Management is a ToolBox developed in the context of a European Union project H2020 (SDK4ED - https://sdk4ed.eu) from University of Macedonia (UoM). It provides analysis in all aspects of technical debt (TD principal, TD interest, TD breaking point, TD interest probability).