| 书目名称 | Safe, Secure, Ethical, Responsible Technologies and Emerging Applications | | 副标题 | First EAI Internatio | | 编辑 | Franklin Tchakounte,Marcellin Atemkeng,Rajeswari P | | 视频video | | | 丛书名称 | Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engi | | 图书封面 |  | | 描述 | This book constitutes the refereed proceedings of the First EAI International Conference on Safe, Secure, Ethical, Responsible Technologies and Emerging Applications, SAFER-TEA 2023, held in Yaoundé, Cameroon, during October 25-27, 2023. .The 24 full papers were carefully reviewed and selected from 75 submissions. They were organized in topical sections as follows: Regulations and Ethics of Artificial Intelligence, Resource-constrained Networks and Cybersecurity, Emerging Artificial Intelligence Applications, Reviews.. | | 出版日期 | Conference proceedings 2024 | | 关键词 | artificial intelligence; internet of things; biases correction; discrimination removal; invasion of priv | | 版次 | 1 | | doi | https://doi.org/10.1007/978-3-031-56396-6 | | isbn_softcover | 978-3-031-56395-9 | | isbn_ebook | 978-3-031-56396-6Series ISSN 1867-8211 Series E-ISSN 1867-822X | | issn_series | 1867-8211 | | copyright | ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2024 |
| 1 |
Front Matter |
|
|
Abstract
|
| 2 |
|
|
|
Abstract
|
| 3 |
Examining Potential Harms of Large Language Models (LLMs) in Africa |
Rehema Baguma,Hajarah Namuwaya,Joyce Nakatumba-Nabende,Qazi Mamunur Rashid |
|
Abstract
Large language models (LLMs) have the potential to generate significant benefits, but their blanket application in Africa could exacerbate existing social and economic inequalities. This is due to a number of factors, including limited technological advancement, historical injustice and marginalization, and underrepresentation of African languages, values, and norms in training data. Despite comprising nearly one-third of the world’s languages, most African languages are underrepresented on the internet: they are primarily oral with little available in written and digitized form. Additionally, most African languages have conflicting orthographic standards. While Africa is undergoing a digital transformation, both internet connectivity and digital literacy remain relatively low and unevenly distributed. This lack of online representation for African languages limits the availability of natural language data for training inclusive language models. This paper examines the potential harms of LLMs in Africa, covering harms already documented for the African context; harms studied and documented for the Western context, but previously unapplied to Africa; and novel potential harms based
|
| 4 |
The Legal Framework of Artificial Intelligence in Cameroon |
Job Nzoh Sangong |
|
Abstract
Artificial Intelligence (AI) represents a system with the capacity to rationally address intricate problems and make appropriate decisions to attain its objectives in various real-world scenarios. Presently, AI finds numerous applications across diverse domains, serving multifaceted purposes driven by a plethora of motivations. However, it is imperative to acknowledge that contemporary AI systems do not possess genuine cognitive intelligence. In light of this, it becomes evident that AI necessitates a legal framework to govern its deployment and operation. This chapter seeks to delve into the legal aspects of artificial intelligence, exploring the legal nature, potential conflicts, and ethical considerations that may arise in the commercial utilization of AI. A central inquiry of this paper revolves around the relationship between artificial intelligence and the law. To address this inquiry, we examine the legal categories into which AI can be classified and, subsequently, the inevitable ramifications that follow such categorizations.
|
| 5 |
|
|
|
Abstract
|
| 6 |
A Gamification Architecture to Enhance Phishing Awareness |
Jean Emmanuel Ntsama,Claude Fachkha,Philippe Brice Owomo,Adrian Chickagwe Focho |
|
Abstract
The development of emerging technologies, such as the Internet of Things and Artificial Intelligence, has provided a spectrum of online and remote solutions in various fields. However, the proliferation of targeted cyberattacks (e.g., phishing) against such technologies has made our assets and data relatively vulnerable to adversaries and hackers. Considering the higher number of victims, it is necessary to evaluate technical solutions that limit the development of a critical mass of people who can participate in collective resistance to such a phenomenon. Raising awareness is, undoubtedly, a way to prevent as many people as possible from falling prey. Given the changes in the different educational theories, we must seek the best way to sensitize cyberspace users to their varied profiles and needs. This study develops an educational gamification architecture that can ensure commitment, motivation, and consideration of a learner’s profile. Subsequently, the problem of the best didactic means is posed with openness to the integration of artificial intelligence, the choice of the type of gamification, and the technologies that can contribute to ensuring that everyone is competent to e
|
| 7 |
Improvement of Cloud-Assisted Identity-Based Anonymous Authentication and Key Agreement Protocol for |
Sidoine Djimnaibeye,Aminata Ngom,Igor Tchappi,Borgou Mahamat Hassan,Amro Najjar |
|
Abstract
Kumar and Chand propose an Identity-based Anonymous Authentication and Key Agreement (IBAAKA) protocol for Wireless Body Area Network in the cloud-based environment, which achieves mutual authentication and user anonymity and can resist known attacks. However, Rakeei and Moazami show that their scheme is subject to a traceability attack. As a result of this attack, the scheme does not allow secure authentication because an adversary can successfully perform a man-in-the-middle attack and exchange a session key with the victim sensor. To provide security resilience against this attack, this paper proposes an improvement to the IBAAKA protocol. We also present the resilience of the proposed scheme against various security attacks, as well as ensuring secure mutual authentication and anonymity.
|
| 8 |
DIDOR: A Decentralized Identifier Based Onion Routing Protocol |
Saha Fobougong Pierre,Mohamed Mejri |
|
Abstract
We propose a new communication protocol, called DIDOR, that provides strong anonymity and is based on a decentralized identifier (DID). The proposed protocol benefits from the persistence, anonymity, and resolvability properties of the DID and integrates a forward secrecy algorithm such as Ephemeral Diffie-Hellman. It is designed to protect the anonymity of communication between parties by combining DID properties and onion routing techniques. The chosen forward secrecy algorithm establishes a shared secret between the sender and the different nodes and strengthens security. The shared secret is then used at each onion layer to encrypt the message. Likewise, DID allows us to offer a protocol that overcomes some of the weaknesses of using digital certificates. Given the prior existence of the decentralized identity management system, our protocol requires less computation at each node, maintains low communication overhead and latency, and does not require directory servers on which most existing protocols rely. We finally analyze the security of our protocol and demonstrate how it can withstand the denial of service attack on DID documents. Overall, this protocol represents a promis
|
| 9 |
Feature Analysis and Classification of Collusive Android App-Pairs Using DBSCAN Clustering Algorithm |
Roger Yiran Mawoh,Franklin Tchakounte,Joan Beri Ali,Claude Fachkha |
|
Abstract
In this study, we tackle the issue of application collusion, which involves multiple apps working together to achieve malicious goals that they couldn’t achieve individually. The current security model of Android, which focuses on permissions, doesn’t effectively address this threat because it only mitigates risks associated with individual apps. To address this limitation, we carried out an extensive analysis of features to identify the key Android permissions utilized by colluding app-pairs. We propose an approach to classify collusive app-pairs using the DBSCAN clustering algorithm. Our results provide valuable insights into the relationship between specific Android permission sets and the malicious activities performed by colluding app-pairs. We identified 12 permissions as the most important features contributing to the classification of collusive app-pairs. We also identified 4 distinct clusters of colluding Android app-pairs.
|
| 10 |
Feature Engineering Considerations in IoT: A Case Study |
Jean-Marie Kuate Fotso,Ismael Abbo,Franklin Tchakounté,William Shu,Claude Fachkha |
|
Abstract
Since the emergence and integration of the Internet of Things, homes have become increasingly intelligent and communicative. These connected homes require special control and security from conception, as they can expose people through confidential data sharing and system attacks. To cope with this, Intrusion Detection Systems remain the best solution, despite the need for improvement and adaptation, since these technologies frequently monitor enormous volumes of data flow with unnecessary and duplicated capabilities, this has a detrimental effect on how well they work. Current work on the Internet of Things shows a real willingness on the part of researchers to propose lightweight, accurate IDS-IoTs with reduced functionality. This study aims to provide an overview of the design of a security solution, by experimenting with Feature Engineering extraction and selection techniques. As such, the PCA, IG, ANOVA, LDA and RFE algorithms were evaluated on the TON_IoT dataset. The features obtained for each technique were evaluated through the Random Forest model using indicators, such as ROC, Accuracy, selection time for each algorithm, and training time for each group of features obtaine
|
| 11 |
Correlation Clustering Adapted for Cell Site Management of Mobile Networks in Developing Countries |
Ado Adamou Abba Ari,Yekoniya Ndjekiltemai,Jocelyn Edinio Zacko Gbadouissa,Arouna Ndam Njoya,Lyse Nao |
|
Abstract
Any mobile network operator’s primary concern is ensuring a better customer experience for their subscribers. For this reason, they need to ensure that their infrastructure is working correctly. However, managing telecommunication infrastructure, especially cellular base stations, has never been an obvious task in the African and Middle Eastern regions due to the landlocked nature and lack of access roads, especially in rural areas. Despite the many solutions developed by operators, ranging from monitoring tools to the deployment of technicians in the field, this still needs to be solved. Some operators prefer to entrust these cell sites to Managed Service Providers (MSPs) or Tower Companies (TowerCos) and concentrate on other services. To address this issue, we propose an adapted correlation clustering for cell site management, considering the operator’s parameters and a site accessibility parameter. This approach makes it possible to determine the optimal number of cells to allocate to a technician to make his interventions efficient; this will minimize Operational Expenditure (OpEx) and cell downtime due to breakdowns and maximize the quality of service offered to customers.
|
| 12 |
A Lightweight Authenticated Key Agreement Scheme for Resource-Constrained Devices Based on Implicit |
Mounirah Djam-Doudou,Ado Adamou Abba Ari,Hortense Boudjou Tchapgnouo,Abdelhak Mourad Gueroui,Alidou |
|
Abstract
In this paper, we discuss the issue of secure communication among devices with limited resources. We introduce a key agreement protocol that utilizes implicit certificates with elliptic curves specifically designed for devices with limited capacity. We establish a certification chain within a finite graph to depict the connection among nodes within the identical group and propose a workload distribution strategy across all cluster nodes. Additionally, we present a trust scheme that enables nodes to generate implicit certificates on an elliptic curve and securely create keys with their counterparts. The group leader acts as the root CA and constructs a hierarchical structure within the finite graph, establishing a certification chain in an organized manner with an intermediate certificate authority (ICA) at every level. This chain is utilized by nodes for generating and sharing implicit certificates, from which symmetric keys for communication between nodes are derived. We then implement the solution using TelosB sensors in the TOSSIM simulator with an AVL Tree. We evaluate the security and resilience of our proposed scheme through informal analysis and a formal model. The informal
|
| 13 |
|
|
|
Abstract
|
| 14 |
Machine and Deep Learning Models for the Prediction of Performance and Speed Regulation Parameters o |
Patrick Njionou Sadjang,Nelson Issondj Banta Jr |
|
Abstract
In this paper, we focus our attention on “Machine Learning and Deep Learning Models for Prediction of Performance and Speed Regulation Parameters of a Turbojet Engine Using Electric Power Transfer Concept”. The principal objective of the study is to implement and compare deep learning and machine learning models for the Prediction of Performance and Speed regulation Parameters of a Turbojet Engine Using the Electric Power Transfer Concept. The novelty of this work is the direct calculation of SFC and Net thrust without any sub-model with a good precision. The data for this study are from the CFM 56–3 turbojet engine equipped with a special EPT architecture. The work showed that the different models (Multi-Linear Regression, Random Forest, and Artificial Neural Networks) give reliable and precise results. Globally neural network model produces the most precise results (Except for LPTCN), and the Linear Regression model is the least precise. The ANN gives an Root Mean Square Error (RMSE) value between 0.19% and 7% of the range of the concerned variable, which is better than those observed in the literature. The results of this work could serve as the first tools for more optimal desi
|
| 15 |
Advancing High-Resolution Weather Prediction Through Machine Learning and GNSS Techniques |
Robert Galatiya Suya |
|
Abstract
Accurate high-resolution weather prediction is essential for numerous applications, including agriculture and disaster management. However, traditional forecasting methods face challenges in achieving precise predictions. This study addresses the challenge of weather prediction by harnessing the fusion of machine learning and Global Navigation Satellite System (GNSS) techniques. Specifically, novel machine-learning models are developed to predict precipitable water vapour and temperature, critical variables affecting weather patterns. The models seamlessly combine GNSS and meteorological datasets from the Continuously Operating Reference Station (CORS) in Mzuzu City, Malawi, to estimate weather attributes with exceptional accuracy. Experimental results demonstrate that the proposed models achieve prediction accuracies, enabling the capture of subtle variations in precipitation and temperature contents. Among the developed models, a suitable candidate for weather forecasting is identified based on its superior performance. The development of this model represents a significant step towards the realization of a real-time weather prediction system capable of providing accurate and loc
|
| 16 |
French-Fulfulde Textless and Cascading Speech Translation: Towards a Dual Architecture |
Tala Metalom Diane Carole,Yenke Blaise Omer,Fendji Kedieng Ebongue Jean Louis |
|
Abstract
Speech-to-Speech translation is attracting increasing attention from researchers due to its potential of easing the communication. It can be leveraged to develop voice user interfaces for important services, such as agriculture in low-literate communities where poorly resourced languages such as Fulfulde are used. If an earlier technique, such as cascading speech, requires textual data, a recent and textless one, such as direct speech-to-speech, only considers speech corpus, playing a crucial role in considering oral languages. In this work, a general approach for a dual architecture is proposed integrating direct speech-to-speech and speech-to-speech translation using automatic speech recognition, machine translation, and text-to-speech translation. Beyond proposing an architecture, this paper focuses on an important step in cascading speech translation, namely automatic speech recognition (speech to text translation). Automatic speech recognition for the Fulfulde language, using the Kaldi toolkit, allowed obtaining an average Word Error Rate of 28.91% for S2T with the monophone acoustic model and a WER of 26.58% for S2T with the triphone acoustic model. The dataset used is based
|
| 17 |
Assessment of Thermal Comfort Using PMV, aPMV, ePMV and TSV Indices in a Naturally Ventilated Buildi |
Tsague Cathy,Medjo Astrid,Jean Seutche,Tchinda Rene |
|
Abstract
Thermal comfort significantly impacts human health and activity in offices, hospitals, and residential and commercial buildings. A study of thermal comfort was carried out in a naturally ventilated building in the city of Yaoundé. Thermal comfort indices such as the predicted mean vote (PMV), the adaptive predicted mean vote (aPMV), the extended predicted mean vote (ePMV) and the thermal sensation vote (TSV) were used for this study by considering two scenarios; the first in an ideal environment and the second in a real environment. Using the Humphrey model, we obtained a comfort temperature of 26.66 °C, corresponding to the operating temperature to the nearest 0.1. Whether in an ideal or real environment, the PMV index is unsuitable for studying thermal comfort in a naturally ventilated building. In an ideal environment, the aPMV index is the most appropriate for assessing thermal comfort, but in a real environment, the most appropriate index is the ePMV index.
|
| 18 |
Classification Analysis of Some Cancer Types Using Machine Learning |
Scott Ulrich Jemea Ebolo,Olusola Samuel Makinde,Berthine Nyunga Mpinda |
|
Abstract
Cancer is a disease caused by changes in deoxyribonucleic acid, which attacks cells in the body, causing them to grow uncontrollably and spread to other parts of the body. Cancer can be deadly. The fact that it can develop anywhere in the body gives rise to many types of cancer. Because a good diagnosis increases the probability of administering a good treatment to save life. Therefore, to reduce the mortality rate from cancer, several diagnostic methods have been developed as the appropriate treatment option is highly dependent on the type of cancer. In this work, we address the issue of classification of some cancer types by using supervised learning methods to classify prostate cancer, lymphoma, leukaemia and small round blue cell tumour. To be more specific, we used five models: support vector machine, decision tree, random forest, K-nearest neighbours (KNN) and artificial neural network. Each cancer dataset was trained using each of the machine learning methods on the Google Colab graphics processing unit (GPU). The test samples were classified for each cancer type, and the performances of the five models were compared in terms of their percentages according to some metrics. T
|
| 19 |
Development of an Intelligent Safety Monitoring Device for Train-Track System in Cameroon |
Tse Sparthan Azoh,Wolfgang Nzie,Bertin Sohfotsing,Tibi Beda |
|
Abstract
Valued as the best land transportation sector when it comes to the movement of heavy tons of freight and plenty of passenger’s at a unique instant, the railway industries have flippantly registered series of accidents that accounted for economic and social deprivation. These accidents directly related to environmental changes, human mistakes and material damages, heavy speeding, overloading, absence of inspection before train departure has portrayed the inconsistency of current maintenance strategies design to guarantee the availability and reliability of traintrack transportation network recently. The important to have an observation platform to preview active and proactive failures in various industries is of great interest. This research paper focuses on suggesting with application a strategy for the design, manufacturing and installation of an intelligent observation tool (IOT) suitable to follow-up failures and repair activities in related industries. Therefore, to guarantee safety of travellers, welfare of freights and protection of infrastructures. However, the strategy uses an inductive technique for failures extraction and a deductive technique for ranking these failures.
|
| 20 |
Towards a Flexible Urbanization Based Approach for Integration and Interoperability in Heterogeneous |
Moskolaï Ngossaha Justin,Ynsufu Ali,Batouré Bamana Apollinaire,Djeumen Rodrigue,Bowong Tsakou Samuel |
|
Abstract
Ensuring seamless integration and interoperability in heterogeneous health information systems is a significant challenge in healthcare. This paper presents a methodological approach that leverages urbanization in information systems to address these challenges. Urbanization involves transforming existing infrastructure incrementally, anticipating constraint changes, and incorporating new technologies. By adopting this approach, healthcare organizations can quickly adapt their systems to meet evolving demands while preserving critical information assets. The study delves into the unique context of health information systems and provides insights into achieving semantic integration. The proposed approach effectively resolves interoperability issues without requiring extensive system reconstruction. It introduces the concept of an Emergence information system, enabling collaboration across diverse contexts, including inter-ministerial, decentralized state services, and non-governmental organizations. Case studies validate the approach, showcasing the benefits of a flexible and urbanized integration strategy. The Yaoundé Urban Information Systems Integration and Interoperability Platf
|
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications影响因子(影响力)
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications影响因子(影响力)学科排名
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications网络公开度
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications网络公开度学科排名
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications被引频次
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications被引频次学科排名
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications年度引用
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications年度引用学科排名
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications读者反馈
书目名称Safe, Secure, Ethical, Responsible Technologies and Emerging Applications读者反馈学科排名
|
|
|
发表于 2025-3-21 17:01:55
