Cryptology and Network Security with Machine Learning

Many of today’s cyberattacks heavily rely on screening addresses over the World Wide Web to find susceptible hardware and software. By keeping track of this scanning activity, you may assess the present situation with regard to several vulnerabilities and their exploitation. Studies examining scanning activity to this point have relied on uninvited traffic obtained from darknets and have concentrated on random screens of the address space. We suggest a method for identifying network scanning threats from both inside and outside the network. Our approach operates on the presumption that every legitimate connection between two computers must typically be accompanied by a DNS resolution; hence, any transfer that is not followed by a DNS inquiry is considered to be an inspection, unless it is permitted by the network’s security protocol. Effective port inspection and sniffing of packet technologies are a must for ensuring strong security. Wiremap, a Python-based port monitoring and packet sniffing programme that makes use of the Scapy library, is introduced in this study. Network administrators, security experts, and penetration testers may evaluate the security posture of their networks using Wiremap’s complete solution. Wiremap’s main goal is to give users a simple, effective tool that lets them scan a target IP address to look for accessible ports and record network data for study. The application takes advantage of the Scapy library’s robust sniffing and packet-dissection capabilities, giving users insights into transmission patterns and possible weaknesses in their networks.

Distributed ledgers, decentralized storage, centralized identity and authentication, and immutable records are just some of the features baked into blockchain technology. It’s moved beyond the hype stage and into practical applications in fields like health care. Public blockchains in the health-care business often need stronger criteria for identification, interoperability, and information sharing than other industries due to regulations such as the repeal of the Health Insurance Portability and Accountability Act of 1996. (HIPAA). Academic and industry researchers have started exploring health-care-related applications that build on existing blockchain technology. Some examples of this include smart contracts and the authentication of identities to avoid fraud. Despite these improvements, concerns remain due to the fact the blockchain presents a unique set of flaws and difficulties that need to be tackled, such as mine incentive, extracting attacks, and key managing. This assessment found, however, that many of the requirements of smart health care were not being met by some of the blockchain initiatives under study. Possible future research directions are also discussed in this study.

This paper discusses various steganography techniques and their parameters. Based on these parameters, we make some comparative analysis among different steganography methods in various research domains. We take LSB, DCT, hash function, and DWT methods for comparisons. The important parameters of the mentioned techniques are PSNR, embedding capacity, and MSE. Based on these parameters, we found that LSB method is most appropriate for implementation of video steganography having negligible error. Video steganography is a mean of hiding videos from one format to another format. The primary objective of video steganography technique is to conceal an important message (information or data) inside of an ordinary video such that an unauthorized person cannot reveal the message.

Hydroponics is the process of growing plants without the use of soil. It is a Latin word which roughly translates to “working water”. Without soil, water provides nutrients, hydration, and oxygen to plant life. Minimal space is used 90% less water than traditional agriculture, and unique design. Hydroponic farms grow fruits and vegetables of excellent quality. Hydroponics still seems to be an under-explored field in India; with our paper, we aim to bring awareness about its benefits and promote its usage. We aim to design and prototype a cost-effective, sustainable, and simplified hydroponic system that will encourage more farmers to take a step towards modernization, along with learning to be more comfortable with new technologies. Our system will be composed of sensors that we shall fit near the water supply source of any hydroponic system so that it can pick up the data from the system and its environment and store it in a cloud database. Our goal is to create a web application utilizing Google Firebase that allows the farmer to see and alter elements such as temperature and pH with a single click in real-time.

By identifying areas of heavy congestion, smart traffic management systems can then implement measures to alleviate that problem. In order to accomplish this, it analyzes and synchronizes data from sensors in real-time. The input data are used to determine when the traffic signal blinks due to congestion. This method has the potential to lessen air pollution and make roads safer for everyone. In this research, we offer a smart technique for accurate traffic estimate prediction using random forest. In the majority of instances, the approximate traffic estimate is better than the shortest route with respect to overall travel time and fuel cost. Based on our findings, we were able to decrease the average waiting time of vehicles while still accurately forecasting traffic.

Sinkhole attacks have been the most dangerous threat to security in recent times. The attackers send malicious requests to Domain Name Server (DNS) that causes unavailability of services and redirects the victim to the destination designed by the malicious attacker. The malicious attacker with the help of these compromised devices forms a BOTNET and uses it for malicious purposes like ransomware, extortion, unauthorized access, fraudulent attempts, data theft, financial gains, eavesdropping, repudiation, etc. The attacker node or the sinkhole node was placed at three different locations to observe the change in count of DIO messages. The purpose of this research was to detect that at what locations can attacker gather a large amount of data. It was observed that when the attacker node was placed at the edge and along the communication links, there was a significant increase in the count of DIO messages. This means that at these points, attacker can gather a lot of information.

The “5G” mobile network is expected to achieve the demanding needs of mobile traffic in the twenty-first century and predicted to be the essential architecture for the advance assistance or services (Wang et al. in Trans Emerg Telecommun Technol 28(9):e3155, 2017). The new services provided by the “5G” network come with new demands and difficulties that cause hindrance in the desirable aim of the advancing mobile network. Mobile communication administrator is again analyzing the design of their network to produce adjustable, changing and effective, commercial and smart solutions. With the emergence of 4G, wireless mobile network took place to provide the services many researchers and cellular network administrators started examining the development for more advance network 5G to fulfill the requirement of higher data rates, greater range of area, lesser delay in providing services and effective QoS (Quality of Service). In the existing network architecture, extensive development is needed to meet the future demands. The key features of 5G wireless cellular network are versatile, convenient and cloud based services which will establish the innovative and futuristic mobile communication as the leading protocol for worldwide communication. In this paper, we talk about the 5G mobile network, about its needs and key features, the drawbacks associated with it and its architecture. We will overview the current technology and shortcoming associated with it and evolution of generation of network with their main characteristic. We will analyze how 5G is different from existing technologies and its impact on society. The 5G network design considerations are also discussed, with cloud radio access network, ultra- dense network, software defined network and network function virtualization examined as key potential solutions towards a green and soft 5G network(Chih-Lin et al. in Philos Trans R Soc A Math Phys Eng Sci 374:20140432, 2016).

IoT devices are physical things that have sensors, network connectivity, and software built into them. This allows them to gather data and share it with other systems and devices. The rising usage of Internet of Things (IoT) systems in critical infrastructure and industrial settings has raised concerns about their vulnerability to cyber-attacks. Therefore, the IIoT desperately needs techniques for enhancing strategic actions. In this work, we propose an Intrusion Detection System (IDS) for IoT devices by using deep learning (DL) and machine learning (ML) algorithms, incorporating Fisher score feature selection on the Edge-IIoT dataset. To develop effective IDS models, we have first applied the Fisher score feature selection method for the identification of the most discriminative features from the Edge-IIoT dataset. For the ML-based IDS, we employ decision tree and random forest models, optimizing their hyperparameters using a systematic hyperparameter tuning approach getting 93.7% (Decision Tree), 94.36% (Random Forest), and 94.5% (Random Forest with hyperparameter tuning) accuracy. For the DL-based IDS, we propose a single-layered feed forward neural network (FFNN) and a multi-layered feed forward neural network (MLFNN) getting 96.1% and 96.5% accuracy, respectively. These models are trained on the selected features from the dataset and evaluated using various performance indicators, including recall, F1-score, accuracy, and precision. Overall, this research work contributes to the field of intrusion detection in IoT devices by combining Fisher score feature selection with both ML and DL algorithms. The findings highlight the prospective of hybrid approaches in increasing the security and resilience of IoT systems, ultimately enabling more robust and efficient intrusion detection mechanisms for IoT deployments in critical domains.

The present research raises few concerns that are arising due to security in cloud computing. As we know several trades are shifting from conventional storage of facts to cloud-enabled storages, it suggests an operative techniques of data access at all times and locations, respectively. Yet, security of the data is one of the foremost concerns, it prevents the establishment to make use of use cloud computing for their existing challenges. A hybrid security system based on some cryptography technique was proposed in this paper. The proposed hybrid model is the amalgamation of techniques based upon symmetric and asymmetric key model. In the present research, the implementation of various techniques of AES and ECC was incorporated which raised the safety of data by incorporating numerous levels of security concerns at both the ends, i.e., sender and receiver, respectively. To lessen the dangers to security of data, the present security standard offers the high level of transparent behavior with respect to users of cloud as well as service providers. The proposed model was implemented by making the use of technologies like Java and the cloud-sim. The cloud-sim and Java were used to implement the suggested model. This methodology increases data security to a greater degree while speeding up text file uploads and downloads in contrast to the current system.

The revolutionary field of natural language processing has broad implications for privacy and safety. This article reviews the wide range of natural language processing uses for privacy protection. Through a systematic literature review, we identify the most important natural language processing approaches employed for various forms of security, such as phishing email detection, cyberthreat analysis, anomaly detection, and privacy-aware text processing. We also discuss the moral issues that must be taken into account while implementing NLP, including defenses against adversarial assaults, good AI protocol, and safeguards for personal data. While examining the possible benefits and hazards of NLP systems, the study emphasizes the significance of responsible and ethical use. The future scope is also discussed to strengthen NLP’s utilization in the domain of privacy and security in data-driven era. This study is of related interest to researchers, practitioners, and policymakers in learning about natural language processing and how it relates to security and privacy.

The human face is the mirror of the mind. The face generally tells all that is going on in one’s heart and mind. Just by looking at the faces of our known ones, we may easily guess their mood. But many times, when we meet some unfamiliar person, it’s hard to get his or her mood just by looking at their faces. This is just because the person may have a certain facial structure that makes them by default look angry, happy, or sad. So, we need to spend some time with that person to analyse other parameters before concluding their state of mood. The current work proposed a novel approach that integrates facial images with electroencephalography (EEG) signals for facial expression recognition tasks. When attention-based deep CNN analyses the facial traits of the subject, a parallel Long Short-Term Memory (LSTM) network analyses the EEG signals. A late fusion network combines the features extracted from both networks, and finally, a classification network tells about what is the current mood of the subject. Combining multiple modalities for emotion recognition has shown promising results when compared with other state-of-the-art models. There are multiple real-life applications of emotion recognition models, such as Advertisement Industry, Human–Robot Interaction, Automatic Depression Detection, Mood Audio/Video Players, etc.

Deceit identification has been a problem since ages and for generations past now. The process is taken in both positive and negative prespectives. Positive for the justice it delivers to the unjustified people and negative for the techniques that are used for the purpose. It has been historically proved that the techniques involved in lie detection (common term used for deceit identification) involve many methodologies including those against human rights and various international conventions. The processes involved have always been in a questions and various studies have tried to prove the deceit; hence, latest methodologies and developments in machine learning area are under trial for such detection jobs. The following paper discusses various supervised learning methodologies like k-Nearest Neighbours, AdaBoost, etc., in order to prove deceit. As general awareness the deceit information used for supervised learning is already labelled and is taken from an open-source dataset. The study tries to establish a basic threshold of parameters in lie detection (LD) and is comparable to the human levels of information gathering with deceit identification with an accuracy of over 70% which is comparable to that of human interrogation techniques. The paper aims to lay a basis for deceit detection using machine learning methodologies which in the future could change the face of LD.

Medical facilities in the different regions of the world are not available as per requirement. More significantly, remote and rural areas have lots of scarcity of quality doctors and diagnosis facilities. Today, the whole world is connected via the Internet which can be used to facilitate medical prescription and diagnosis at distant places. Everything about these facilities is defined by a hypernym that is telemedicine. Medical data of the patients is very confidential and can be misused if someone is able to access the network and retrieve the data. Security mechanisms, such as digital watermarking and other primitive cryptographic technologies, are in use, but these technologies have their limitations. Therefore, there is a need to propose a security mechanism that can provide confidentiality, integrity, and authentication. The research proposes a security mechanism based on blockchain. The security features provided by blockchain, along with cryptographic primitives, can be used to safeguard the information. Most of the medical data is in an image format. The pixels of the data are arranged along with the patient's personal information, and a corresponding hash value is generated. Every terminal in the blockchain network is updated with the basic agreement and a hash value. It is protected with a proof-of-work mechanism which is very complex even to brute force. Along with this, elliptic curve digital signature algorithm and partial homomorphic encryption algorithm are also implemented to further ensure confidentiality, integrity, and authenticity of the data. The paper also presents a use case where we have used visual cryptography as a means to secure the medical images.

Payment gateways are necessary for securing online transactions on e-commerce platforms, but there is still a problem with the restricted use of cryptocurrency as a payment method. Due to lengthy confirmation periods and fees, scalability concerns on blockchain networks like Bitcoin and Ethereum prevent the implementation of micropayments. With the help of a method of payment that makes use of the benefits of the Polygon distributed ledger network for quick and affordable transactions, we suggest a method of payment that gets beyond these restrictions in this article. The MetaMask electronic wallet securely manages user transaction data, enabling privacy and independence. The solution aims to encourage wider utilization of cryptocurrencies in online shopping by combining blockchain innovations with electronic wallets. The figures obtained demonstrate the advantages that retailers can have as a result.

In recent years, there has been a significant surge in interest in WSNs among researchers and the general public. WSNs are purposefully designed to cater to a broad spectrum of applications, ranging from compact healthcare surveillance systems to extensive environmental monitoring projects. The WSN ecosystem encompasses a multitude of sensor nodes/devices that interconnect billions of diverse objects via the Internet. These sensors nodes are predominantly low-energy devices engineered for intermittent or continuous transmission. Consequently, the significance of energy efficiency (EE) in WSNs cannot be overstated. Driven by this imperative, conserving energy in such systems to extend their lifetime has been the subject of significant research. The design and architecture of sensor nodes play a significant role in energy efficiency. In this paper, we will talk about various optimization techniques that can be used to improve their performance. These include selection and comparison of low-power components, Dynamic Voltage Scaling, energy harvesting, and optimizing the power supply. By focusing on node design considerations, this paper provides insights into of diverse sensor node design optimization techniques tailored to enhance energy efficiency within WSNs. Explored the trifold areas of energy consumption, hardware optimization strategies, and advancements in energy harvesting, elaborating their roles in advancing energy efficiency within WSNs.

Cardiovascular issues have now become the most common reason for the mortality of diabetic and hypertensive patients. Evaluating various risk factors for cardiovascular complexities has become significantly important for the prevention of these issues. Many medical professionals make cardiovascular complications diagnoses on the basis of prior information and data related to various decision parameters. The manuscript provides a methodology for the assessment of risk measures for cardiovascular complexities with the incorporation of the q-rung picture fuzzy decision-making technique. A new methodology involving the q-rung picture fuzzy set up has been proposed and implemented in the diagnoses of cardiovascular complexities. The assessment of risk factors has been done on the basis of proposed methodology for the betterment of the patients.

Hong et al. presented an attribute-based combined public key signature scheme for wireless body area networks. According to the authors, their scheme enables the signer to maintain confidentiality, unforgeability, privacy, and collusion resistance. In the scheme, encryption and digital signatures are combined. Threshold policy is used in the scheme as the access policy. We review and study this work. We observe that the correctness of the verification is not given in the security analysis of this scheme. Correctness of the verification is a fundamental aspect that ensures the validity and reliability of the scheme. We comprehensively present the correctness of verification of this attribute-based combined public key signature scheme for wireless body area networks.

Wireless sensor network (WSN) is an interconnection of numerous sensor nodes with autonomous processing and communication capability. Sensor nodes collect data from the surroundings and transmit it to an endpoint receiver which is known as a sink, so that appropriate decisions can take place. WSN has resource constraints in terms of bandwidth, energy, storage, and processing. Due to limited resources, congestion may occur in WSN. Congestion is one of the serious issues in the network which is degrading the performance of the network and quality of service (QoS). Some factors like packet collision, channel contention, and data transmission in many-to-one fashions become the catalyst for congestion and result in packet drop. Many researchers provided different solutions to solve the issue, but congestion is still a serious problem. Therefore, more specialized strategies are required to manage congestion. In this paper, we have investigated various congestion detection and management methods and compared them based on queue, priority, source management, traffic management, packet service time, channel load, and buffer occupancy.

With the popularity of blockchain innovation, data security has become a concern. The blockchain framework is transparent and immutable, but its unique strategy creates problems for protecting sensitive data. This article discusses attribute-based encryption (ABE) as a plan to improve data security in the context of blockchain. ABE concept in the context of blockchain provides a good way to ensure data security, improve customer privacy, and increase trust in a distribution system. This paper examines the building blocks of ABE and VANET and examines its potential application to protect information security within the blockchain framework. Additionally, the paper highlights the importance of integrating privacy-enhancing advances like VANET into blockchain systems to address evolving challenges in data protection. By taking advantage of attribute-based encryption, the blockchain framework can reduce the security risks associated with the current while maintaining the immutability and security key led by blockchain innovation.

COVID-19 pandemic has changed plenty of things around us. Now days, the use of face mask is very essential for protection against the deadly virus, and at some places, it is compulsory too. Similarly, temperature measurement of people specifically in public places like school, colleges, cinema halls, and some other crowded areas becomes a necessity, but it is a very tedious task to manage. It has also been observed that people do not wear face mask putting the life of others in danger. It is not possible to keep an eye on everyone in crowded places, and manually, it is next to impossible. By analyzing this problem, in this paper, an IoT-based face mask and temperature detection system using Arduino is presented. The measurement of temperature is performed using MLX90614 temperature sensor which sends an alert signal if the value exceeds the pre-determined threshold value. The face mask detection is done by a Convolutional Neural Network (CNN)-based face mask classifier. Additionally, there is one more system present in which if the temperature of a person is found above the threshold value, then in that case, a sanitizer dispenser will not dispense a sanitizer and that person will not be allowed to enter into the premises. So, the proposed method is simple, economical, and robust enough to be used in crowded places and can be implemented anywhere and very effectively which eases our work.

Modern world moves around data communication, and its security is a key issue nowadays. It is very unsafe to transmit images over insecure wireless communication channels. In this paper, a secure and reliable image encryption method using RSA is proposed. RSA is one of the most trusted and reliable public key encryption algorithms used over the years. We use the strongness of RSA for transmitting images over insecure channels and provide an easy implementation technique. The proposed method is capable to encrypt any image using RSA parameters (selected by user). Initially, a public–private key pair is generated, and then, the image (which is to be transmitted) is encrypted by the public key where as private key is used for the decryption process. The most important point associated with the proposed method is that it gives freedom to users to select their own parameters making the method convenient, customized, and robust and provides high level of security.

Fully Homomorphic Encryption (FHE) schemes enable secure computations on encrypted data. Following Gentry’s groundbreaking result, the AGCD problem-based FHE scheme, also known as Fully Homomorphic Encryption over the Integers (FHE-OI), was introduced by Dijk et al. Over the time, several improvements have been made to FHE-OI, including the CS scheme proposed by Cheon and Stehlé. This paper presents two significant enhancements to CS scheme. The first contribution involves extending their FHE scheme to support message space $$\mathbb {Z}_g$$ Z g , removing the previous constraint limited to binary numbers as described in Cheon and Stehlé’s work. Building upon this advancement, the second enhancement further extends the scheme to encompass batch fully homomorphic encryption. This extension empowers the scheme to efficiently encrypt and perform homomorphic operations on entire vectors of plaintext bits using a single ciphertext, thereby enhancing its applicability and utility in various practical scenarios.

Designated verifier signatures (DVSs) represent a specific type of digital signature that holds significance exclusively for a designated verifier. This feature emphasizes the importance of the intended receiver authenticating the signature, rendering it impenetrable to verification by any other party. DVSs provide a number of potential benefits over standard digital signatures, including increased security, flexibility, and efficiency. This paper conducts a thorough examination of cutting-edge research in the field of DVSs. It digs into the many types of DVSs that have been proposed, including ID-based DVSs, proxy DVSs, designated verifier strong signatures, bi-designated signatures, bi-designated proxy signatures, designated proxy threshold signatures, and multi-proxy signatures and exploration of the several applications that make use of DVS capabilities. In addition, we also propose DVS proxy signature scheme, secure against Type 5 and Type 6 challenger attack.

The extensive and growing utilization of the Internet and mobile apps has resulted in the enlargement of the online realm, rendering it more vulnerable to extended and automated cyber assaults. In response to this heightened vulnerability, cybersecurity techniques have been developed to strengthen security measures and improve the ability to detect and respond to cyberattacks. Due to the intelligence of cybercriminals in evading traditional security systems, the previously employed security measures have become inadequate. Conventional security systems struggle to effectively detect new and ever-changing security attacks that are previously unseen or have varying forms. ML methods are making substantial contributions to different aspects of cybersecurity, playing a pivotal role in numerous applications within the discipline. While ML systems have been successful so far, there are considerable obstacles in ensuring their trustworthiness. This paper’s main objective is to offer a thorough examination of the obstacles ML techniques encounter in safeguarding cyberspace from attacks. This is accomplished by examining the existing body of literature concerning ML techniques utilized in the field of cybersecurity. These techniques encompass areas such as intrusion detection, spam detection, and malware detection within computer and mobile networks. The document also provides succinct elucidations of each specific machine learning approach, indispensable machine learning tools, ML involvement in cybersecurity, and current state of ML for cybersecurity. Finally, the paper examines the barriers and challenges, as well as the anticipated path for the future of ML in the context of cybersecurity.

In this article, we have proposed a new concept of a complex picture hesitant fuzzy set which has advantages over the two theories of picture fuzzy set and hesitant fuzzy set. Later, some algebraic properties related to the proposed theory have also been explained in detail which adds the strong foundation of the concept for further study. An application based on the theory has also been presented which validates the applicability of the concept in solving the problems caused by the fuzziness of the data available.

The creation of a blockchain website for student certificate verification is a project designed to answer rising concerns about the legitimacy and security of academic credentials. This project intends to establish a trustworthy and efficient platform for certifying student credentials by exploiting the decentralized and tamper-resistant properties of blockchain technology. The project's goals include creating a transparent, fraud-resistant system that protects the integrity of certificate data. Sensitive student information will be encrypted and shielded against unauthorized access using cryptographic methods and smart contracts. The blockchain website will simplify the verification process, allowing educational institutions, future employers, and other stakeholders to readily validate students’ credentials. Implementing a blockchain website for student credential verification provides increased trust, security, and efficiency. The blockchain maintains the integrity and dependability of certificates by removing the potential of manipulation or fraud. The simplified verification procedure saves time and resources for all parties involved, enhancing overall efficiency. However, difficulties like as scalability, interoperability, user experience, and legal frameworks must be overcome for the project to be implemented successfully. Finally, the deployment of a blockchain website for student certificate verification is a viable approach for combating certificate fraud and increasing trust in academic credentials. This project’s goal is to leverage blockchain technology to build a secure, transparent, and user-friendly platform that revolutionizes the verification process and assures the integrity of student credentials.

The integration of Mobile Ad hoc Networks and the Internet of Things offers innovative applications but also presents significant security challenges. This paper explores these challenges, considering the decentralized nature of both MANETs and IoT. Security vulnerabilities, including unauthorized access and malicious attacks, emerge due to factors like node mobility and resource constraints. Existing security mechanisms are examined and found insufficient for this combined environment. The primary security issues are confidentially, integrity, availability, blackhole, wormhole, grayhole, rushing, flooding, attacks, authentication, intrusion detection, and trust management tailored to MANET-IoT systems. The urgent need for comprehensive security strategies to ensure safe operation of this ecosystem is emphasized.

During the past few years, especially after the emergence of the Covid-19 pandemic, researchers have devoted their efforts in improving the global health sector by supporting it with the latest technologies. Among these technologies, we often hear about Internet of Medical Things (IoMT) and blockchain in the effort of facilitating the patients and medical staff to preserve their security and confidentiality of the patient data and protect it from every hacking attempt to steal or falsify. In fact, during just the last three years alone, dozens of new schemes have been proposed in the literature for the integration of blockchain with IoMT for the healthcare field. Therefore, we present in this paper a review of some the notable works in this area. Our intent is to explain the basic principles of this field and classify the notable proposed schemes, as this study suggests potentially interesting avenues for future research to use it as a reference material by the researchers in this field.

Communication is critical for power system operation and control. Data transfer via digital means is vulnerable to security and privacy concerns. Man-in-the-middle (MITM) attacks pose a risk to the data integrity of synchrophasor values transferred as digital data within the smart grid. The purpose of this research is to demonstrate how utilizing Media Access Control (MAC) spoofing instead of Address Resolution Protocol (ARP) spoofing can improve the stealth and robustness of MITM attacks in synchrophasor networks and to provide a thorough understanding of the weaknesses and possible risks present in synchrophasor networks when they are the target of MITM attacks, highlighting network elements and targets. After a session hijack, when bogus data was eventually injected, it was evaluated and processed by the phasor data concentrator (PDC), ensuring consistent networking and the attacker's non-detectability. For generic experiment results, different sub-networks based on the master–slave topology of synchrophasor networking and commercially available phasor measurement unit (PMU) and PDC were used in a laboratory-scale setup.

The real estate industry holds a crucial position within global economies, exerting considerable impact on a range of sectors including steel, cement, and banking. In the context of India, the service sector holds the position of being the second-largest employment, following agriculture. This growth is primarily attributed to the expanding urban middle class. Over the course of the last decade, there has been a notable and remarkable upswing in the real estate industry, mostly driven by a heightened need for both commercial and residential properties. Metropolitan areas such as Mumbai, the National Capital Region (NCR), Bangalore, and other similar urban centers, in conjunction with their surrounding suburban regions, constitute significant economic hubs. In light of the pandemic-induced challenges, the real estate industry has experienced a resurgence in conjunction with the broader economic recuperation, driven by post-pandemic population movements, and a revitalized sense of trust among investors. Significantly, there has been a notable increase of 87% in private property investment during the past two years, indicating a revived sense of confidence in the market. India’s prospective role as a manufacturing hub in the aftermath of the COVID-19 pandemic has garnered global attention, owing to forecasts of significant investments in real estate and infrastructure. Despite the presence of favorable opportunities, there are nevertheless persistent obstacles. This study investigates the challenges encountered by the Indian real estate industry and investigates the possibility of blockchain technology in addressing these challenges.

Securing image data during communication influences employing encryption, authentication, integrity reviews, and other protection measures to stop unauthorized access and tampering. Encryption methods contain confusion and diffusion to preserve the security of image data by intricately changing the relationship among original and encrypted data and distributing pixel effects across the entirety of the encrypted image. Therefore, by prioritizing security while also addressing computational resources, we have designed a system for encrypting images. This includes using the row and column index manipulator algorithm for initial scrambling and integrating the split and join algorithm for an added coating of confusion. To accomplish diffusion goals, a pseudo-random sequence generator is created, utilizing the logistic map for both diffusion and confusion processes. The results and estimation of the proposed technique highlight its strong randomness and resilience against entropy-based, statistical, differential, and brute-force attacks.

In this article, we design some public-key encryption schemes using various frameworks like McEliece, Niederreiter, HyMES, and a new framework proposed by Ivanov et al. taking Hadamard code as secret code. We provide key generation, encryption, and decryption algorithms and also give toy examples to illustrate these algorithms. We discuss various attacks and also implement all these encryption schemes in SageMath software.

This research paper explores the vulnerabilities of the lightweight block cipher SPECK 32/64 through the application of differential analysis and deep learning techniques. The primary objectives of the study are to investigate the cipher’s weaknesses and to compare the effectiveness of ResNet as used by Aron Gohr at Crypto2019 and DenseNet. The methodology involves conducting an analysis of differential characteristics to identify potential weaknesses in the cipher’s structure. Experimental results and analysis demonstrate the efficacy of both approaches in compromising the security of SPECK 32/64.

In 2018, He et al. worked on Chang and Le’s authentication method for ad hoc wireless sensor networks and found that it undergoes problems related to security. Based on their study of Chang and Le’s protocol, He et al. designed a scheme that can withstand various types of security threats. In this article, we analyze He et al.’s scheme and find that it has a few issues related to security. He et al.’s method cannot withstand the attack named as an insider attack; an adversary can disrupt the session key establishment; it does not support the validation of the owner of the smart card before initiating the authentication process; and lacks mutual authentication. To remove these problems from He et al.’s protocol, we present an enhanced authentication method for ad hoc wireless sensor networks.

Future payment methods will include cryptocurrencies. Bitcoin was peer-to-peer, much like BitTorrent. A multiparty signature examines the ledger against the transaction of each pulse. In general, a deal where people distrust each other, but don’t have identities, comes into force. Cryptocurrencies were created with the primary goal of developing a distributed transaction system that allows for the settlement of concurrent legal contracts. To ensure double spending is avoided, every money made goes to only one recipient. Libra was launched on June 18, 2019, as a cryptocurrency. At the same time, Facebook introduced Calibra, an electronic wallet. Despite this, financial institutions have come out forcefully against the plan since it puts the sovereignty of a country at risk. This piece will outline the project, detailing all of its features and how they affect our society. In response to the regulatory pressures, the Libra Association announced a rebranding of the project to “Diem” in late 2020. Diem aimed to address concerns by focusing on a simplified and scaled-down version of the original concept. The project’s scope shifted to focus on providing a digital payment solution for existing financial systems, working in alignment with regulatory frameworks.

The increasing computational power has greatly empowered deep learning algorithms, making the creation of highly realistic and virtually undetectable synthetic videos, commonly known as deep fakes, remarkably simple. These deep fakes, often featuring convincing face swaps, can be employed in situations such as generating political turmoil, fabricating terrorism events, distributing revenge porn, and engaging in blackmail. In this study, we introduce an innovative deep learning-based technique designed to effectively differentiate between AI-generated fake videos and genuine ones. Our approach excels at automatically identifying manipulated videos, including those involving content replacement and reenactment deep fakes. Our endeavor involves leveraging Artificial Intelligence (AI) to combat its own creations. The central component of our system utilizes a ResNext Convolutional Neural Network (CNN) to extract features from individual frames of videos. These features are then used to train a Long Short-Term Memory (LSTM)-based Recurrent Neural Network (RNN). This RNN helps classify videos by determining if they have been manipulated, such as being deepfakes, or if they are genuine recordings. To ensure our model performs well with real-time data and reflects real-world situations, we evaluate it on a large and balanced dataset created by combining various existing datasets like FaceForensic++, Deepfake Detection Challenge, and Celeb-DF. Furthermore, we showcase that our system achieves competitive results using a straightforward yet effective approach.

Virtually all organizations have embraced cloud computing to differing extents in their operations. However, along with this adoption of cloud technology, it’s imperative to ensure that the organization’s cloud security strategy is equipped to safeguard against the primary threats to cloud security. The largest worry with cloud computing is security, which also prevents many consumers from using these services. The security concerns at various cloud infrastructure layers are presented in-depth in this article. We delve into the attributes and methods of delivery presented by cloud computing, along with the conceivable hurdles and limitations encountered during its integration into the business environment. Our analysis extends to the cybersecurity aspects associated with cloud computing, with a specific focus on intrusion detection and prevention mechanisms, as well as their applicability within the cloud context. In this research paper, we have implemented the Google Cloud Platform for security considerations. Paper also discusses about provider and customer responsibility ratio for better security of cloud environment.

Face recognition technology using AI has seen paradigm shift in the evolving world. Automatic attendance monitoring using real-time face identification is a solution to handle attendance in any small/large as well as scalable organization. Traditional methods in the organization involve calling names or signing sheets with individuals, which is a very time-consuming process and provides insurance. This is also subjected to manual errors. Automation of attendance recording and monitoring through face recognition is a process of identifying the face for taking attendance by using the image of the human face as biometric parameter captured through a surveillance camera in the premise. This article presents an effective way of attendance monitoring by making use of deep learning technology and compares its results with the state-of-the-art approaches.

It is certainly peculiar that even after securing one's infrastructure with state of the art technologies, companies still get compromised. The question thus arises ‘How is an attacker able to circumvent such sophisticated defenses?’. The answer to this is relatively simple. Attackers exploit the most vulnerable component in the chain, also known as humans. They do so by targeting people with fake emails and websites. Attackers often spoof legitimate services and modify them to perform nefarious activities. This act of portraying a malicious resource as a legitimate resource is known as phishing. The main motive behind phishing is to trick the victim into revealing personal information or more often phishing acts as a precursor to malware infections. Advancement in technology has made it easier for attackers to spoof a legitimate resource with almost zero flaws. It makes it extremely difficult for the victims to evade such attacks. However with the aid of artificial intelligence detecting such websites becomes extremely easy and accurate. In this research we propose a hybrid deep learning model to detect phishing websites. The hybrid model is a combination of CNN and RNN algorithms and gives a high degree of accuracy in phishing website detection. For training and validation the datasets have been used. The results of our experiments show that the proposed model performs better than traditional deep learning models.

In the rapidly evolving realm of cyber-security, the detection of network anomalies serves as a pivotal line of defense against a myriad of malicious activities and cyberthreats. This research undertakes the task of enhancing the accuracy and efficacy of network anomaly detection by employing a comparative analysis of various individual machine learning models. The study delves into the performance of distinct models, including Random Forest, Gradient Boosting, AdaBoost, neural networks, and SVM, meticulously scrutinizing their capabilities in detecting network anomalies. The crux of this study lies in its meticulous evaluation of each individual model on the revered NSL-KDD dataset—an established benchmark within the field of network intrusion detection. Through a systematic blend of rigorous mathematical frameworks, precise model implementations, and comprehensive experimental assessments, this research offers a deep understanding of the inner workings of each algorithm. The pivotal aspect of this study revolves around the comprehensive comparative analysis of these standalone models. Going beyond the mere quantifica-tion of accuracy, the exploration delves into aspects of precision, recall, $$F_1$$ F 1 -score, and more, shedding light on their diverse facets of performance. With achieved accuracies of 99.2419% for Random Forest, 99.5197% for Gradient Boosting, 86.6044% for AdaBoost, 84.00% for neural networks, and 87.00% for SVM, this research underlines the distinctive attributes and potential of each model in the context of network anomaly detection. As the study unravels the distinct strengths and limitations of each model, it contributes to the broader landscape of cyber-security by providing insights into the efficacy of individual machine learning approaches.

Nowadays, biometric attendance system is applicable in various organizations to ensure the person's identity. The system takes parameters such as person's voice, fingerprint, retina, and face to authenticate the person. In order to track students’ daily attendance when they enroll in college, this article provides a biometric attendance system that uses blockchain technology. Both the daily and cumulative attendance of each student over the course of a chosen period of time, or month, is displayed by this system. The proposed system is designed to provide the cost-effective, systematic, and feasible solution to keep real-time attendance of the students against the existing manual system. Also, a webpage is designed for the same to provide a user-friendly and flexible platform to the students and faculty both to check the student's attendance.

Secure text transfer is a process of securely transferring text-based data over a shared network, ensuring that the data remains confidential and private and that no third party can access or intercept it. One standard method of secure text transfer is through the use of encryption. Several encryption algorithms, such as advanced encryption standard (AES) and Rivest–Shamir–Adleman (RSA), provide varying levels of security. Another way of secure text transfer is through protocols such as hypertext transfer protocol secure (HTTPS), encrypting data transmitted between a web browser and a server, ensuring that attackers cannot intercept it. Similarly, secure file transfer protocol (SFTP) is a secure version of FTP that encrypts data in transit. Overall, secure text transfer is essential when sensitive information is transmitted over a network. Organizations can help ensure their data remains safe and secure by implementing appropriate security measures. There are many places/applications where people use this technique to transfer data securely.

This paper focuses on developing a web application using the Express framework and Node.js. MongoDB Atlas cloud version is used for data storage, with AWS cloud providing the necessary infrastructure. The application integrates Google OAuth for user authentication, ensuring secure and reliable access. The storybook application is designed to be intuitive and easy to use, with a simple and clean user interface. The application also features social sharing functionality, allowing users to share their stories on various social media platforms and connect with others who may be interested in their work. Modern web development technologies and secure authentication protocols ensure a seamless and enjoyable user experience. With the ability to upload stories in public or private mode and the option to connect with others who share similar interests, this application will surely be a hit among writers and enthusiasts alike. To ensure that our application provides a superior user experience, we conducted thorough research on the needs and preferences of writers and storytelling enthusiasts. We found that existing platforms lacked options for users to share their stories securely and privately. Our application addresses this problem by allowing users to upload stories publicly or privately, giving them complete control over who can access their content. In terms of performance, we have tested the application using metrics such as accuracy, precision, and recall. Our results show that the application performs at high levels of accuracy and reliability, ensuring a seamless user experience. Overall, our application provides a solution to the limitations of conventional online platforms, offering writers and storytelling enthusiasts a secure and user-friendly platform to share their stories online.

Biometric surveillance systems, combined with artificial intelligence (AI) and soft biometrics, have revolutionized the identification, and tracking of individuals based on their unique physical and behavioral characteristics. However, the integration of these technologies raises important ethical considerations and necessitates robust legal frameworks. This paper explores the amalgamation of AI, soft biometrics, and human surveillance, focusing on the ethical implications and legal regulations governing their use. The paper provides an overview of biometric surveillance systems, explaining the components and types of biometric modalities employed, such as fingerprints, facial recognition, iris recognition, voice recognition, and gait analysis. It highlights the applications and benefits of biometric surveillance systems in various domains, including law enforcement, access control, and health care. Ethical considerations in the deployment of biometric surveillance systems are thoroughly examined, including privacy concerns, informed consent, discrimination, and bias, and ensuring accountability and responsible use of technology. The significance of robust legal frameworks is emphasized, and an analysis of existing laws and regulations in different jurisdictions, including the UN, USA, Europe, France, Japan, India, Australia, and New Zealand, is provided.

This paper presents an overview of setting up a Virtual Private Network (VPN) using OpenVPN on the Google Cloud Platform (GCP). With remote work becoming increasingly popular, VPNs provide a secure and reliable solution for employees to access cloud resources from any location. OpenVPN is an open-source VPN solution that offers advanced security features, making it a preferred choice for businesses and organizations. Google Cloud Platform provides a secure and scalable infrastructure for deploying a VPN, including global load balancing and network security features. The article outlines the deployment of OpenVPN Access Server on a GCP virtual machine instance and the configuration of user authentication and authorization, network settings, and security policies. Creating a virtual machine instance is the first step in setting up a VPN using OpenVPN on GCP. The OpenVPN Access Server is then deployed on the virtual machine instance, and user authentication and authorization protocols are configured. Network settings such as IP address range, routing, and firewall rules must also be configured to ensure that only authorized traffic can pass through the VPN. Testing and monitoring the VPN infrastructure is essential to ensure optimal performance and reliability. This involves testing VPN connectivity from remote clients and monitoring network traffic and server logs for unusual activities. However, setting up a VPN using OpenVPN on GCP provides a secure and scalable infrastructure for remote access to cloud resources. With the outlined steps, organizations can deploy a VPN infrastructure quickly and efficiently, with advanced security features and the flexibility to scale as their needs evolve.

Cloud computing environment combines a variety of networked devices to facilitate the provision of requested services. 8–9% challenges related to cloud are due to identity and access management. Numerous academics and industry experts have diligently tackled the challenges associated with ensuring secure access to cloud resources. It constitutes a core element of cloud security and assumes a pivotal role in the protection of data and assets housed and handled within cloud ecosystems. Identity management is used to authorization of registered users or customers and access management is used to provide access to user of data and services provided by cloud. Various identity management and access management protocols are compared in this research paper. Need of Zero Trust Architecture has been analyzed. Two major access management schemes are implemented with 10 roles and 18 polices using Zero Trust Architecture and its results shows the improvement in security performance of cloud environment.

Recently, the fusion of industrial processes and communication technologies has emerged as Industry 4.0. In this emergence, smart devices such as sensors and mobile phones and machines have become data prosumers (i.e., producer and consumer both simultaneously). The exchange of data between these prosumers takes place using the public Internet. In this system, generally, producers send colossal data to consumers via a cloud server. So that, the consumer can easily access the required data. As prosumers are generally resource-constrained smart devices (like sensors, actuators, etc.), therefore, the processing and analysis of such massive data are done by the cloud servers. Thus, the receiver device gets processed data. However, sometimes the response is delayed. To overcome the latency, cloud computing and fog computing have been merged together so that the time-bound data may be processed on the fog node, as it is near to prosumers. However, due to public connection, the security and privacy issues like data leakage/modification or illegal admittance are major problems. To banish such issues, a certificateless aggregated signcryption scheme has been proposed by Dohare et al. (published in IEEE Transactions on Industrial Informatics, doi: 10.1109/TII.2022.3142306). However, after a detailed analysis, we found various flaws in their scheme. In this paper, we present the cryptanalysis of the Dohare et al.’s scheme. To show the cryptanalysis effectively, we present a detailed discussion by pondering the mathematical and theoretical concepts.

The paper explores the challenges posed by ransomware attacks in the healthcare sector within the context of the digital transformation of healthcare. It examines real-world incidents, such as those at AIIMS hospital, to highlight the disruptive nature of ransomware attacks and underscores the importance of proactive defense strategies. The study introduces three innovative approaches to ransomware prevention in healthcare: blockchain, machine learning, and software-defined networking (SDN). Each approach is analyzed in terms of its role in safeguarding healthcare data. Blockchain ensures data integrity and access control through decentralization, while machine learning enhances threat detection by identifying unusual behaviors, potentially indicative of ransomware. SDN provides dynamic network segmentation, real-time responses, and centralized security updates to counteract attacks. The paper concludes by summarizing the benefits and challenges associated with these methods and emphasizes the necessity of collaboration among healthcare professionals, technologists, and policymakers for effective implementation. These innovations are crucial for the healthcare industry to navigate the evolving cybersecurity landscape and safeguard patient data.

IoTs are being very popular these days due to their application in a number of areas. However, they are also used in some sensitive applications where security is a prime concern like military operations, whereby if the network is compromised then the outcomes can be disastrous. Secure routing is a challenging task in mobile ad hoc networks where no fixed infrastructure exists. Trust can be a solution to handle the soft security intimidation. The identities on the IoT networks should be verified using an effective trust management method to ensure secure and reliable routing. However, the nodes feature and task complexity perform managing trust a difficult process. The proposed trust-based proposed approach includes various dimensions of trust such as direct trust, behavioral, and recommendation in trusting with the gap filling purpose along with dishonest recommendation for neighbors and energy efficiency. This paper proposed a secured route selection approach for IoT environment by using very popular Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). Moreover, Intuitionistic fuzzy is applied to manage not only numerical trust value but linguistic data also. The results of experiments specify that the proposed framework utilize the parameters under consideration. The current work in addition, compute the total trust by considering all three dimensions The trusted value of indirect nature is derived with the help of trust value indicated by the recommending nodes based on their trustworthiness with the evaluating, and the direct trust value is determined with the supporting of existing contacts between the evaluating and evaluated. The direct trust value is reliable and resistant to untruthful recommendations. The dishonest suggestions of the other nearby nodes, however, can make an indirect recommendation vulnerable. The use of intuitionistic fuzzy approach can solve the problem of dishonest advice. Finally trust values are used in data maintained as well moving toward trust value trust value for each neighbors node to opt for the secured path between two nodes.

The prediction of encryption keys in cryptographic systems poses a significant security threat. In this study, we investigate the feasibility of predicting the keystream used in the Espresso stream cipher through the implementation of LSTM and CNN architectures. The Espresso stream cipher, known for its lightweight design and cryptographic strength, serves as the basis for our analysis. By leveraging the capabilities of LSTM and CNN, we aim to assess the vulnerability of the Espresso stream cipher to keystream prediction attacks. By training and evaluating LSTM and CNN models on a dataset consisting of keystream bytes, we examine the extent to which the keystream used in the Espresso stream cipher can be predicted. Our experimental results demonstrate the effectiveness of LSTM and CNN in predicting the keystream bits or bytes of the Espresso stream cipher. By shedding light on the weaknesses of the Espresso stream cipher in the context of keystream prediction, this research work contributes to the field of cryptographic security.

With the rising usage of cloud services, data centers (DC) are improving the services to their customers. The substantial energy consumption (EC) of cloud DCs poses significant economic and environmental challenges. To address this issue, server consolidation through virtualization technology has emerged as a widely adopted approach to decrease energy consumption rates, minimize virtual machine (VM) migration, and prevent breaches of service-level agreements (SLAs) within data centers. Cloud DCs are becoming larger, consuming more energy, and capable of delivering quality of service (QoS) with service-level assurance. People all around the world can use cloud computing to have instant access to resources. It provides pay-per-use services via a vast network of data center locations. The data centers that house cloud servers are kept operational to provide a variety of services, which uses a lot of electricity and has an adverse environmental impact. The primary goal of cloud computing is to offer uninterrupted and continuous Internet-based services, while using virtualization technologies to satisfy end users’ QoS requirements. With the balanced EC and service quality, it is challenging to supply cloud services. The rapid expansion of cloud services significantly rises energy and power consumption daily. This paper reviews previous studies on multiple parameters such as EC, SLA violation, and VM migration by different approaches based on statistical techniques, machine learning approaches, heuristic, and metaheuristic methods. Prediction of host CPU, identifying underload or overload hosts, VM consolidation have been applied to manage the resources using the PlanetLab and Bitbrains workload on different performance metrics. This review paper presents a detailed comparative study of different algorithms to analyze the influence of several parameters such as energy consumption, SLAV, virtual machine migration, active hosts, etc. on the performance of cloud resources. As a result, effective VM consolidation reduces power consumption, VM migration, and SLA assurance during service provisioning. It has been found that the statistical methods save up to 28% of energy, 90% SLAV, and 90% VM migration. The machine learning-based method reduces energy consumption up to 45%, SLAV up to 63%, VM migration up to 50%, the heuristic approaches save up to 72% energy, 78% SLAV, 46% VM migration, and the metaheuristic methods reduce 25% energy consumption, 79% SLAV, 89% VM migration compared to the related benchmark methods for a variety of parameters and configurations.

Predictive models are employed in order to forecast forthcoming events of which knowledge is now lacking. This is achieved by analysing a collection of pertinent predictors or variables, taking into account both contemporary and past data. Predictive modelling, alternatively referred to as predictive analytics, encompasses the utilization of statistical methodologies, data mining techniques, and artificial intelligence approaches to address a diverse range of applications. In the field of healthcare, a predictive model is utilized to acquire knowledge from past patient data in order to forecast future medical issues and subsequently select the most appropriate course of therapy. This review emphasizes the application of deep learning (DL) models, including LSTM-Bi-LSTM, RNN, CNN, RBM, and GRU, in various healthcare contexts. The findings suggest that the LSTM/Bi-LSTM model is commonly employed in the analysis of time-series medical data, whereas CNN is frequently utilized for the examination of medical picture data. The utilization of a model based on deep learning has the potential to support healthcare personnel in expediting decision-making processes pertaining to prescriptions and hospitalizations, resulting in time savings and enhanced service provision within the healthcare business; especially for COVID-19 case, this model can be used which makes it better suitable for examining medical images. The present study examines the many prediction models employed in healthcare applications through the utilization of deep learning techniques and also makes sure that the extracted dataset is secured while processing it.

The utilization of CT scans in medical diagnostics has seen a consistent and substantial rise. However, this increased usage has raised concerns regarding the potential harmful effects of radiation exposure on patients. Reducing the radiation dose can result in more noise in the captured images, which can negatively impact the radiologist's ability to make accurate judgments with confidence. The most commonly encountered types of noise in medical images include Gaussian noise, speckle noise, and salt and pepper noise. Numerous significant efforts have been made to enhance image quality by eliminating this noise, and deep learning-based methods have gained popularity due to their effectiveness in handling various types of noise and image datasets. Within the research community, various neural network variations, such as autoencoders, generative adversarial networks (GANs), residual networks, convolutional neural networks (CNNs), and regularized neural networks, have gained immense popularity. In this paper, we comprehensively discuss eleven highly impactful approaches for image denoising based on deep learning techniques. We assess the performance of these methods using two quantitative and effective metrics: structural SIMilarity (SSIM) and peak signal-to-noise ratio (PSNR).

The Reserve Bank of India passed A Circular dated 6 April 2018 (the ‘circular’) following that, the Indian Parliament formed an Inter-Ministerial Committee in 2019 which formulated the draft of Banning of Cryptocurrency and Regulation of Official Digital Currency Bill, 2019. Both the Circular and Bill of 2019 have been stringent to an extent that they’ve interfered with the fundamental rights of the citizens of India. Where the RBI through its Circular banned the banks and legal entities from entering into any relationship with firms dealing with crypto-assets, the Draft Bill completely banned the use and possession of digital currency. The authors have analysed the Draft Bill and the landmark judgement of Internet and Mobile Assn. of India versus Reserve Bank of India, pronounced by the Supreme Court of India. The authors have also dived into the technological aspect of the crypto-assets and tried to understand the how this technology evolved.

Stock price prediction has long been a pivotal area of interest for investors, financial analysts, and researchers alike. The ability to forecast future stock prices accurately can provide substantial benefits in investment decision-making. With the advent of machine learning techniques and the ever-increasing availability of financial data, the field of stock price prediction has witnessed significant advancements. This paper presents a comprehensive review of stock price prediction methods using machine learning approaches. The primary objective is to provide an in-depth analysis of the various techniques, their strengths, limitations, and their overall performance in the context of stock market forecasting. Stock price prediction is categorizing into three main groups—(1) Statistical Models which include traditional time series models like ARIMA and GARCH. (2) Supervised Learning Models which explores the application of regression, decision trees, support vector machines, and various ensemble methods for stock price prediction. (3) Deep Learning Models like recurrent neural networks (RNNs), LSTM, and convolutional neural networks (CNNs). To evaluate the effectiveness of these methods, we review empirical studies and compare their performance on real-world stock market datasets. The paper concludes by summarizing the key findings, identifying challenges that still need to be addressed, and highlighting potential future directions in this dynamic field. This comprehensive review aims to serve as a valuable resource for researchers, practitioners, and investors interested in leveraging machine learning techniques for stock price prediction, shedding light on the current state of the art, and guiding future research endeavors.

In this paper, we will examine numerous optimization approaches in the field of computer science engineering in depth, shedding light on their applications, strengths, and weaknesses. Optimization algorithms are important tools in computer science engineering, with applications spanning from machine learning to computer vision, data mining, robotics, and more. In principle, optimization algorithms strive to locate the best possible solution among a group of possibilities while taking certain objectives and restrictions into account. They are the foundation of problem-solving approaches, providing a systematic and efficient approach to dealing with multiple difficulties. The efficiency and efficacy of each algorithm vary from one another, and each algorithm has advantages and limits that rely on the applications they are used with. We intend to provide a comprehensive view of optimization algorithms. We will cover their many types, delving into their real-world applications and painstakingly analyzing their strengths and weaknesses. In addition, we will investigate the complexities of each algorithm, giving light on the specific characteristics and settings in which they shine. This work seeks to serve as a basic resource for computer science engineering academics and practitioners, developing a deeper understanding of optimization algorithms and stimulating more inquiry in this dynamic field.

The diagnosis and detection of numerous diseases has advanced significantly in the healthcare sector, which is always changing. One illness that has significantly impacted humankind is diabetes, a condition that directly impacts blood glucose levels. Glucose, or sugar, is the primary source of energy for our bodies, and it is derived from the food we consume. Insulin, produced by the pancreas, assists glucose in entering the cells of the body. But diabetics are either unable to use their own insulin well or do not create enough of it, resulting in increased levels of carbohydrates in the body. New diseases are being diagnosed at an alarming rate, which is indicative of the impact that changes in our lifestyle habits have had on our health. This paper is about fulfilling two major objectives, i.e. (i) The dataset has been made secured by applying encryption generation key to it. This will help in maintaining the privacy of the patients and also will avoid unauthorized access. (ii) Secondly, in order to predict diabetic retinopathy in the patient’s various machine learning models have been used. This work truly shows the importance of data security and data preservation using cryptography (Fernet). It can help clinicians in making better decisions during critical stages of treatment. Our findings show how well machine learning and data security operate to diagnose diabetic retinopathy, and they also point to areas that could be improved in the future with the use of deep learning models and frameworks.

The utilization of key agreement protocols within the domain of cryptography is indispensable for ensuring secure and confidential communication across untrusted networks. This research paper offers a comprehensive exploration of the significance and application of key agreement protocols in modern cryptographic systems. Covering classical techniques like Diffie–Hellman and more recent advancements such as Elliptic Curve Diffie–Hellman, the paper examines their mathematical foundations and operational strengths. The security aspects of key agreement protocols are scrutinized, encompassing potential threats and vulnerabilities. The paper underscores the importance of thwarting attacks like man-in-the-middle, while emphasizing the role of forward secrecy and authentication mechanisms. Moreover, the paper investigates the adaptability of key agreement protocols in diverse scenarios, ranging from Internet communication to resource-constrained environments like IoT devices. By presenting a comparative analysis of various protocols, considering security, performance, and applicability, the paper aids decision-making for protocol selection. This research contributes to a comprehensive understanding of key agreement protocols, enabling practitioners to make informed choices in establishing resilient cryptographic channels amid evolving cybersecurity challenges.