Papers by Valentin Sgarciu
New mechanisms for privacy in human-in-the-loop cyber-physical systems
Nowadays, we witness a tremendous increase in systems that sense various facets of humans and the... more Nowadays, we witness a tremendous increase in systems that sense various facets of humans and their surrounding environments. In particular, the detection of human emotions can lead to emotionally-aware applications that use this information to help improving people's daily lives and offering new business opportunities. We address the issue of maintaining privacy for this type of applications. In this paper, we propose a general model that focuses on privacy-preserving mechanisms for a Human-in-the-loop emotionally-aware Cyber-Physical System (HiTLCPS). As a proof-of-concept, we also present an emotionally aware application that attempts to positively impact students' lives without compromising their privacy1.
The need for quality improvement, along with the possibility of disassembling products for enviro... more The need for quality improvement, along with the possibility of disassembling products for environmental protection implies storing of a product history, starting from development and manufacturing phase until it reaches the end of its lifecycle. Based on possibility of a smart tag attachment to a product and going along with the idea of decentralizing an assembly line, the paper herein proposes a system which can be implemented at every operational point within an assembly line.
The paper presents an application developed for Java Card smart cards, with multiapplication supp... more The paper presents an application developed for Java Card smart cards, with multiapplication support. Although the multiapplication support was not exemplified, the use of smart card for personal identification has some advantages, the most important beeing its security. In this paper are shown the structure of the application, various security systems, which ensure the access to a objective to only the designated people, who have rights to enter in the certain objective.
Data Mining Techniques for Employee Evaluation
The article shows how complex the human resource field is, how much artificial intelligence can i... more The article shows how complex the human resource field is, how much artificial intelligence can influence a field and how some companies already used artificial intelligence (AI) to improve their human resources (HR) capabilities. The paper’s main objective was to identify a way to evaluate whether the managers promote or offer benefits to employees in an objective way. In the study, three data mining algorithms were evaluated in order to determine their level of efficiency when applied on the available data set.
Comparative Study on Feature Selection Methods Rooted in Swarm Intelligence for Intrusion Detection
Today, the cyber-space landscape is filled with vulnerabilities and malicious users that exploit ... more Today, the cyber-space landscape is filled with vulnerabilities and malicious users that exploit them. Adding to this the technological context, with data flooding from all directions, it is evident that maintaining a good level of security for our systems has become a challenge. The high-dimensional datasets generated by all these diverse technologies needs to be processed efficiently. Inside the pool of data records gathered from all the monitored devices, there is also noise that could negatively influence intrusion detection. Feature selection methods(FSM) may offer a plausible solution to this multi-dimensionality problem, because they reduce the number of features and keep only those significant for solving the stated problem. In this paper we propose to conduct a comparative study of feature selection methods for intrusion detection. We focus on wrapper variants of FSM which are based on swarm intelligence algorithms. To conduct our study, we construct our FSM models based on four SI algorithms (PSO, BA, BAL and BAE) in combination with traditional classifiers (SVM, C4.5 and Naive Bayes) and use the NSL-KDD dataset for our test and comparative analysis.
Smart refrigerator: A next generation refrigerator connected to the IoT
The Internet of Things (IoT) refers to the set of devices and systems that interconnect real worl... more The Internet of Things (IoT) refers to the set of devices and systems that interconnect real world sensors and actuators to the Internet. This includes many different systems, including smart objects, smart monitoring devices, home automation systems, smartphones (which are increasingly being used to measure the world around them) and many more. This paper presents the concept, architecture, building process and functionality of such a device that incorporates the systems stated above: a refrigerator that is exposed as an IoT object and interacts with the items stored within, gathers information about them, process this information into relevant data that is later conveyed through an IoT platform to its owners, in other words, a smart refrigerator.
Wireless system for occupancy modelling and prediction in smart buildings
The dense instrumentation of future smart buildings enables the implementation of advanced contro... more The dense instrumentation of future smart buildings enables the implementation of advanced control techniques which are aimed at the dual objectives of energy efficiency/cost savings and occupant comfort. One of the essential functions and prerequisite consists of robust dynamic occupancy detection and prediction which allows improved estimation of active thermal zones and internal loads, as compared to static schedule-based approaches. The paper presents the design and preliminary evaluation of a wireless system for embedded monitoring of occupancy states across the thermal zones of the building. The challenges of calibration, detection and prediction algorithms are discussed along with experimental outcomes. The infrared array sensor within the system offers improved detection performance compared to conventional PIR sensors while preserving user privacy in comparison to image processing approaches using security cameras. Several advantages of the proposed solution are also highlighted such as the low cost, flexibility, scalability and integration towards the building-wide automation system.
Anomaly Intrusions Detection Based on Support Vector Machines with an Improved Bat Algorithm
The continuous proliferation of more complex and various security threats leads to the conclusion... more The continuous proliferation of more complex and various security threats leads to the conclusion that new solutions are required. Intrusion Detection Systems can be a pertinent solution because they can deal with the large data volumes of logs gathered from the multitude of systems and can even identify new types of attacks if based on anomaly detection. In this paper we propose an IDS model which includes two stages: feature selection with information gain and detection with Support Vector Machines (SVM). A draw-back of SVM is that its performance results are influenced by its user input parameters. Therefore, in order to better the classifier we exploit the advantages of a recent Swarm Intelligence (SI) algorithm, the Bat Algorithm (BA), which we improve by enhancing its randomization with Lévy flights. We test our model for the NSL-KDD dataset and prove that it can outperform the original BA, ABC or the popular PSO.
Indoor positioning using RFID and REST
This paper presents the concept of real-time positioning aligned with the Web of Things paradigm ... more This paper presents the concept of real-time positioning aligned with the Web of Things paradigm in order to create a decoupled cloud service that can be easily integrated, scaled and adapted to any type of objects being tracked or positioned. A proof of concept of the proposed system was developed in order to demonstrate and evaluate both its architecture and its features.
IFAC Proceedings Volumes, Oct 1, 2003
Product traceability during its development phases was a preoccupation of engineers and people wh... more Product traceability during its development phases was a preoccupation of engineers and people who made researches in robotics domain. Today, with the help of integrated technology, chips can be attached to units from their starting phases. Therefore, big assembly lines can be decentralized; therefore they do not need to communicate in every moment with a central database. In this paper it is presented a potential integrated system, based on programmable chips and reader in every important point of an assembly line. It is also proposed a simulation model, based on contact less smart cards, based on the fact that a smart card chip can be assimilated with a programmable chip.
LED Smart Illumination with RFID Indoor Positioning
There are already plenty of Internet of Things (IoT) architectures and systems available worldwid... more There are already plenty of Internet of Things (IoT) architectures and systems available worldwide and this creates challenges for software development and systems entities that use these platforms. Global visibility and interoperability through the exposure of these platforms via the web can be achieved through a new paradigm to support an open framework for Web of Things (WoT), which is analogous to web pages. Thus, real-world objects and cloud services could interact via the web. This paper presents the concept of an intelligent lighting system designed in an architectural style based on the Representational State Transfer (REST) web technology and provides intelligent illumination based on Radio-Frequency Identification (RFID) technology.
IFAC Proceedings Volumes, Oct 1, 2003
This paper focuses upon the diagnosis of an assembling/dissembling robot, using analytical method... more This paper focuses upon the diagnosis of an assembling/dissembling robot, using analytical methods based on observers. The paper presents a modality of detecting incidents, using the mathematical model of the robot, which is based upon the analytical evaluation of the residue with adaptive limits.
IFAC Proceedings Volumes, Oct 1, 2003
The paper presents an application developed for Java Card smart cards, with multiapplication supp... more The paper presents an application developed for Java Card smart cards, with multiapplication support. Although the multiapplication support was not exemplified, the use of smart card for personal identification has some advantages, the most important beeing its security. In this paper are shown the structure of the application, various security systems, which ensure the access to a robot mainframe to only the designated people, who have rights to control the specified robot.
IPv6 Sensor Networks Modeling for Security and Communication Evaluation
Studies in systems, decision and control, 2016
Ubiquitous monitoring and control in future internet-connected pervasive systems architectures re... more Ubiquitous monitoring and control in future internet-connected pervasive systems architectures rely increasingly on Wireless Sensor Networks (WSN) as enabling technology. This can be achieved through seamless integration along existing networking devices and systems, while taking account the data-driven nature and security constraints imposed upon such low power, efficient embedded networked platforms. We argue that only through a combined security and communication approach can the relevant goals be achieved, by considering security issues as an integral part of high level layers of the protocol stack. This chapter presents an extended analysis of security and communication constraints of IPv6-based wireless sensor networks at the network and transport layer. Leveraging the Contiki operating system for embedded sensing nodes along with the ContikiSec security layer and RPL, IPv6 Protocol for Low Power and Lossy Networks. The results of the analysis are discussed and functions of its modules are explained with focus on future reliable and large scale implementations.
An Improved Bat Algorithm Driven by Support Vector Machines for Intrusion Detection
Springer eBooks, 2015
Today, the never-ending stream of security threats requires new security solutions capable to dea... more Today, the never-ending stream of security threats requires new security solutions capable to deal with large data volumes and high speed network connections in real-time. Intrusion Detection Systems are an omnipresent component of most security systems and may offer a viable answer. In this paper we propose a network anomaly IDS which merges the Support Vector Machines classifier with an improved version of the Bat Algorithm (BA). We use the Binary version of the Swarm Intelligence algorithm to construct a wrapper feature selection method and the standard version to elect the input parameters for SVM. Tests with the NSL-KDD dataset empirically prove our proposed model outperforms simple SVM or similar approaches based on PSO and BA, in terms of attack detection rate and false alarm rate generated after fewer number of iterations.
On Using a Cloud-Based Approach to Develop a Mobile Asset Management Solution
Studies in systems, decision and control, 2016
After the 9/11 events, stakeholders around the world have become concerned and looked for solutio... more After the 9/11 events, stakeholders around the world have become concerned and looked for solutions to address the safety of the global transportation infrastructure. The Mobile Assets Management System (MAMS) presented in this paper focuses on improving transportation safety related aspects such as vehicle safety, cargo safety, passenger safety and driver safety. MAMS is a mobile asset management solution dedicated to locating and monitoring commercial fleets, offering fleet managers and dispatchers complete monitoring and control capabilities from an Internet-connected computer located anywhere in the world. The system is built on a cloud based secure, intelligent, interlinked and interactive infrastructure, namely the S4I I™ platform developed by SafeMobile. The implementation of the MAMS application is a reliable proof of the flexibility and the scalability of the S4I platform, truly creating a unique cloud-based integration system for wireless radio communication. We present the context and the benefits of the mobile asset management solution, the architecture and the main components of the system and we underline the advantages it provides to the global transportation infrastructure in terms of safety and security, which is a major concern nowadays.
Current Web search engines return long lists of ranked documents that users are forced to sift th... more Current Web search engines return long lists of ranked documents that users are forced to sift through to find relevant documents. This paper introduces a new approach for clustering Web search results, based on the notion of clustering by compression. Compression algorithms allow defining a similarity measure based on the degree of common information. Classification methods allow clustering similar data without any previous knowledge. The clustering by compression procedure is based on a parameter-free, universal, similarity distance, the normalized compression distance or NCD, computed from the lengths of compressed data files. Our goal is to apply the clustering by compression algorithm in order to cluster the documents returned by a Web search engine in response to a user query.
Studies in Informatics and Control, Jun 15, 2015
This paper focuses on the compression based clustering and aims to determine the most suitable co... more This paper focuses on the compression based clustering and aims to determine the most suitable combinations of algorithms for different clustering contexts (text, heterogeneous data, Web pages, metadata and so on) and establish whether using compression with traditional clustering methods leads to better performance. In this context, we propose an integrated cluster analysis test platform, called EasyClustering, which incorporates two subsystems: a clustering component and a cluster validity expert system, which automatically determines the quality of a clustering solution by computing the FScore value. The experimental results are focused on two main directions: determining the best approach for compression based clustering in terms of context, compression algorithms and clustering algorithms, and validating the functionality of the cluster analysis expert system for determining the quality of the clustering solutions. After conducting a set of 324 clustering tests, we concluded that compressing the input when using traditional clustering methods increases the quality of the clustering solutions, leading to results comparable to the NCD and the cluster analysis expert system proved 100% its accuracy so far, so we estimate that, even if some slight deviation should occur, it will be minimal.
Mining Data for Human Resources
2019 11th International Conference on Electronics, Computers and Artificial Intelligence (ECAI), 2019
The article shows the complexity of the human resource field, offering an insight on the issues t... more The article shows the complexity of the human resource field, offering an insight on the issues that many decision makers are facing, in a world where any software solution designed to offer value, is no longer viable without big data capabilities. This paper's main objective is to prove the importance of mining human resource data by proving its accuracy. Furthermore, such a way of mining big data can be a starting point for more intelligent systems that could offer decision makers valuable real-time information. The study shows an objective view on how accurate the public available web search data is and how it can be mined for achieving valuable information for HR.
This paper presents an approach to integrate underwater sensor networks and a proposed common fra... more This paper presents an approach to integrate underwater sensor networks and a proposed common framework model for multi-purpose devices to be deployed around offshore oil and gas drilling exploration areas, as well as on other types of concrete or steel structures available at sea. The study is focused on underwater sensor devices called “Safe-Nets” developed within the VMAX PerrySlingsby ROV Simulator environment. Furthermore, we are looking into environmental safety and applications in order to justify the costs of these underwater data collection modular devices and we address the deployment challenges using remote operated vehicles (ROV) in scenarios done on the ROV Simulator.
Uploads
Papers by Valentin Sgarciu