Sepideh Kalateh

Dr. Sepideh Kalateh

Electrical and Computer Engineering, NOVA School of Science and Technology, NOVA University Lisbon, Caparica, Portugal
+1
orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
18 July 2024 in IEEE Access
Verified
Journal Article
Open access
Sepideh KalatehSepideh Kalateh
Sepideh Kalateh
LELuis A. Estrada-Jimenez
Luis A. Estrada-Jimenez
Sanaz Nikghadam-HojjatiSanaz Nikghadam-Hojjati
Sanaz Nikghadam-Hojjati
Jose BarataJose Barata
Jose Barata
Emotion recognition involves accurately interpreting human emotions from various sources and modalities, including questionnaires, verbal, and physiological signals. With its broad applications in affective computing, computational creativity, human-robot interactions, and market research, the field has seen a surge in interest in recent years. This paper presents a systematic review of multimodal emotion recognition (mer) techniques developed from 2014 to 2024, encompassing verbal, physiological signals, facial, body gesture, and speech as well as emerging methods like sketches emotion recognition. The review explores various emotion models, distinguishing between emotions, feelings, sentiments, and moods, along with human emotional expression, categorized in both artistic and non-verbal ways. It also discusses the background of automated emotion recognition systems and introduces seven criteria for evaluating modalities alongside a current state analysis of mer, drawn from the human-centric perspective of this field. By selecting the PRISMA guidelines and carefully analyzing 45 selected articles, this review provides comprehensive perspectives into existing studies, datasets, technical approaches, identified gaps, and future directions in MER. It also highlights existing challenges and potential applications of the MER.


7 July 2025 in Sustainability
Verified
Journal Article
Open access
PDF file
Sanaz Nikghadam-HojjatiSanaz Nikghadam-Hojjati
Sanaz Nikghadam-Hojjati
Eda MarchettiEda Marchetti
Eda Marchetti
MGMaria Gustavsson
Maria Gustavsson
FFFilipa Ferrada
Filipa Ferrada
Ana Inês OliveiraAna Inês Oliveira
Ana Inês Oliveira
Laura AndreicaLaura Andreica
Laura Andreica
The demands and preferences of men are often prioritized in societal challenges, despite women comprising half of the world’s population. Gender-Responsive Research and Innovation (GRRI) offers an approach to address this bias, promoting comprehensive and equitable solutions to better-understood real-world problems. The successful implementation of GRRI depends on recognizing obstacles, staying informed about relevant initiatives, and using this knowledge to develop effective strategies, ultimately contributing to a fairer and more responsive society. This research, using a mixed-methods approach, identified 10 critical GRRI concerns, measures, and strategies to address them, as well as 18 noteworthy initiatives across research, industrial, and regulatory contextual fields. It also enhances the understanding of GRRI, laying the groundwork for an ontology of GRRI that can benefit policymakers, organizations, and researchers seeking more inclusive strategies and practices.

25 June 2023
Verified
Book Chapter
Open access
PDF file
LELuis A. Estrada-Jimenez
Luis A. Estrada-Jimenez
Sepideh KalatehSepideh Kalateh
Sepideh Kalateh
Sanaz Nikghadam-HojjatiSanaz Nikghadam-Hojjati
Sanaz Nikghadam-Hojjati
JBJose Barata
Jose Barata
Biologicalisation defines the analysis of biological patterns as a source of inspiration to model intelligent manufacturing systems. Due to their inherent adaptability, these design representations are highly desirable considering the increasing complexity of modeling current manufacturing solutions. Contributing to the idea of self-organizing and autonomous shop floors, we present a framework that aims to support the collaborative healing of manufacturing resources. This has been inspired on the altruistic behavior of bats in which social care is demonstrated (e.g., in bat colonies) even at the cost of individuals’ fitness/health. These ideas are conceptually showcased in an emergent automation manufacturing application, i.e., peer-to-peer energy sharing in automated guided vehicles. Some conclusions, and potential future research are discussed at the end of the paper.

19 December 2025 in Informatics
Verified
Journal Article
Open access
PDF file
NFNastaran Farhadighalati
Nastaran Farhadighalati
LELuis A. Estrada-Jimenez
Luis A. Estrada-Jimenez
Sepideh KalatehSepideh Kalateh
Sepideh Kalateh
Sanaz Nikghadam-HojjatiSanaz Nikghadam-Hojjati
Sanaz Nikghadam-Hojjati
Jose BarataJose Barata
Jose Barata
Healthcare faces a critical challenge: protecting sensitive medical data while enabling necessary clinical access. Evolving user behaviors, dynamic clinical contexts, and strict regulatory requirements demand adaptive access control mechanisms. Despite strict regulations, healthcare remains the most breached industry, consistently facing severe security risks related to unauthorized access. Traditional access control models cannot handle contextual variations, detect credential compromise, or provide transparent decision rationales. To address this, SAFE-GUARD (Semantic Access Control Framework Employing Generative User Assessment and Rule Decisions) is proposed as a two-layer framework that combines behavioral analysis with policy enforcement. The Behavioral Analysis Layer uses Retrieval-Augmented Generation (RAG) to detect contextual anomalies by comparing current requests against historical patterns. The Rule-Based Policy Evaluation Layer independently validates organizational procedures and regulatory requirements. Access is granted only when behavioral consistency and both organizational and regulatory policies are satisfied. We evaluate SAFE-GUARD using simulated healthcare scenarios with three LLMs (GPT-4o, Claude 3.5 Sonnet, and Gemini 2.5 Flash) achieving an anomaly detection accuracy of 95.2%, 94.1%, and 91.3%, respectively. The framework effectively identifies both compromised credentials and insider misuse by detecting deviations from established behavioral patterns, significantly outperforming conventional RBAC and ABAC approaches that rely solely on static rules.