Irina Tal

Dr. Irina Tal


Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 September 2018 in ACM Computing Surveys
Verified
Journal Article
Open access
PDF file
ACAlexandra Covaci
Alexandra Covaci
LZLonghao Zou
Longhao Zou
ITIrina Tal
Irina Tal
GMGabriel-Miro Muntean
Gabriel-Miro Muntean
Gheorghita GhineaGheorghita Ghinea
Gheorghita Ghinea
Mulsemedia—multiple sensorial media—makes possible the inclusion of layered sensory stimulation and interaction through multiple sensory channels. The recent upsurge in technology and wearables provides mulsemedia researchers a vehicle for potentially boundless choice. However, in order to build systems that integrate various senses, there are still some issues that need to be addressed. This review deals with mulsemedia topics that remain insufficiently explored by previous work, with a focus on the multi-multi (multiple media-multiple senses) perspective, where multiple types of media engage multiple senses. Moreover, it addresses the evolution of previously identified challenges in this area and formulates new exploration directions.

27 July 2023 in Information
Verified
Journal Article
Open access
PDF file
MFMichael Mayowa Farayola
Michael Mayowa Farayola
ITIrina Tal
Irina Tal
RCRegina Connolly
Regina Connolly
Takfarinas SaberTakfarinas Saber
Takfarinas Saber
Malika BendechacheMalika Bendechache
Malika Bendechache
Artificial Intelligence (AI) can be very beneficial in the criminal justice system for predicting the risk of recidivism. AI provides unrivalled high computing power, speed, and accuracy; all harnessed to strengthen the efficiency in predicting convicted individuals who may be on the verge of recommitting a crime. The application of AI models for predicting recidivism has brought positive effects by minimizing the possible re-occurrence of crime. However, the question remains of whether criminal justice system stakeholders can trust AI systems regarding fairness, transparency, privacy and data protection, consistency, societal well-being, and accountability when predicting convicted individuals’ possible risk of recidivism. These are all requirements for a trustworthy AI. This paper conducted a systematic literature review examining trust and the different requirements for trustworthy AI applied to predicting the risks of recidivism. Based on this review, we identified current challenges and future directions regarding applying AI models to predict the risk of recidivism. In addition, this paper provides a comprehensive framework of trustworthy AI for predicting the risk of recidivism.

6 December 2017 in ACM Computing Surveys
Verified
Journal Article
Open access
PDF file
ITIrina Tal
Irina Tal
GMGabriel-Miro Muntean
Gabriel-Miro Muntean
Vehicular networks and their associated technologies enable an extremely varied plethora of applications and therefore attract increasing attention from a wide audience. However, vehicular networks also have many challenges that arise mainly due to their dynamic and complex environment. Fuzzy Logic, known for its ability to deal with complexity, imprecision, and model non-deterministic problems, is a very promising technology for use in such a dynamic and complex context. This article presents the first comprehensive survey of research on Fuzzy Logic approaches in the context of vehicular networks, and provides fundamental information which enables readers to design their own Fuzzy Logic systems in this context. As such, this article describes the Fuzzy Logic concepts with emphasis on their implementation in vehicular networks, includes classification and thorough analysis of the Fuzzy Logic-based solutions in vehicular networks, and discusses how Fuzzy Logic could be employed in the context of some of the key research directions in the 5G-enabled vehicular networks.

26 April 2024 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Open access
MFMohammad Fardad
Mohammad Fardad
GMGabriel-Miro Muntean
Gabriel-Miro Muntean
ITIrina Tal
Irina Tal
Vehicular Edge Computing (VEC) has emerged as a promising paradigm to enable low-latency Vehicle-to-Everything (V2X) services by bringing computing resources closer to vehicles. However, the high dynamicity of vehicular networks poses significant challenges in designing an optimal policy for delivering V2X services while ensuring security and timely service delivery. To address these challenges, this paper proposes a BlockchainEnabled Vehicular Edge Computing (BEVEC) framework that employs a dual-layer verification process empowered with a permissioned blockchain to ensure data accuracy and integrity. A novel system utility function is designed to measure the performance of the BEVEC, which also serves as the basis for a consensus mechanism of the permissioned blockchain. To optimize this utility, a Deep Reinforcement Learning (DRL) algorithm is proposed to enable timely service delivery in BEVEC. Simulation-based results demonstrate the effectiveness of the proposed algorithm when compared to existing approaches. On average, it obtained an 18% reduction in latency, a 38% improvement in successful service delivery, and a 65% decrease in energy consumption.

29 August 2023
Verified
Conference Paper
Open access
MFMichael Mayowa Farayola
Michael Mayowa Farayola
ITIrina Tal
Irina Tal
Malika BendechacheMalika Bendechache
Malika Bendechache
Takfarinas SaberTakfarinas Saber
Takfarinas Saber
RCRegina Connolly
Regina Connolly
Artificial Intelligence (AI) is applied in almost every public sector because of its positive impacts. However, AI’s ethical aspects and trustworthiness constitute a significant uproar and concern among different AI stakeholders due to AI’s adverse effect on users when the AI system lacks cautionary measures. AI is used in the criminal justice system for predicting recidivism risk. However, AI’s negative impact translates into bias and high incarceration towards a group of defendants in a population assessed for recidivism risk. This paper focuses on fairness as a requirement of a trustworthy AI framework previously proposed to ascertain the appropriate application of AI systems in predicting recidivism. This paper aims to raise awareness about the fairness of AI models and stimulate further research and deployment of efficient and effective exploitation of fair and trustworthy AI models in the criminal justice system when predicting recidivism. Fairness has been a significant concern for criminal justice system stakeholders and has received considerable attention with more theoretical and practical studies than other trustworthy AI requirements. Hence, this paper reviews state-of-the-art fairness, outlines valuable findings, and proposes future directions to achieve fair AI systems for predicting recidivism risk. In addition, this paper ensures mapping existing technical works in the literature to the fairness pipeline corresponding to the criminal justice system’s AI development phases.

15 July 2021 in IEEE Security & Privacy
Verified
Journal Article
Closed access
RTRamona Trestian
Ramona Trestian
GXGuodong Xie
Guodong Xie
Pintu LoharPintu Lohar
Pintu Lohar
ECEdoardo Celeste
Edoardo Celeste
Malika BendechacheMalika Bendechache
Malika Bendechache
ITIrina Tal
Irina Tal
We introduce a study examining people?s privacy concerns during COVID-19 and reflect on people's willingness to share their personal data in the interest of controlling the spread of the virus and saving lives.