International Journal of Systematic Innovation

A comparative analysis for deep learning-based approaches for image forgery detection

Ravikumar ch, Marepalli Radha, Maragoni Mahendar, Pinnapureddy Manasa — Thu, 22 Feb 2024 00:00:00 +0000

The detection of counterfeit photographs is critical in the digital age because of the widespread development of digital media and its significant impact on social networks. The legitimacy of digital content is being threatened by the growing sophistication of picture counterfeiting. With the help of pre-trained VGG-16 models and deep learning techniques that integrate Error Level Analysis (ELA) and Convolutional Neural Networks (CNNs), this study presents a fresh solution to this problem. The study thoroughly assesses and contrasts these models with a dataset that has been carefully chosen in order to bring the presented findings in perspective. To ensure a reliable evaluation of each model's performance 5000 experiments were carried out in total. With an accuracy rate of 99.87% and an accurate identification rate of 99% of hidden forgeries, the results demonstrate the exceptional effectiveness of the ELA-CNN model. However, despite its robustness, the VGG-16 model only achieves a significantly lower accuracy rate of 97.93% and a validation rate of 75.87%. This study clarifies the relevance of deep learning in the identification of image forgeries and highlights the practical ramifications of various models. Moreover, the research recognizes its constraints, especially for highly advanced counterfeits, and proposes possible paths for enhancing the accuracy and scope of detection algorithms. In the ever-changing world of digital media, the thorough comparative analysis provided in this study offers insightful information that can direct the creation of accurate forgery detection tools, protecting digital content integrity and reducing the effects of image manipulation.

An improved self-training model to detect fake news categories using multi-class classification of unlabeled data

Oumaima Stitini, Soulaimane Kaloun, Omar Bencharef, Sara Qassimi — Thu, 22 Feb 2024 00:00:00 +0000

In recent times, significant attention has been devoted to classifying news content in academic and industrial settings. Some studies have focused on distinguishing between fake and real news using labeled data and have achieved some success in detection. Digital misinformation or fake news content spreads through online social communities via shares, re-shares, and re-posts. Social media has faced several challenges in combating the distribution of fake news information. Social media platforms and blogs have become widely used daily sources of information due to their low cost and ease of access. However, this widespread use of social media for news consumption has led to the dissemination of fake news, creating a severe problem that adversely affects individuals and society. Consequently, identifying and addressing misinformation has become an essential and critical task. Detecting fake news is an emerging research area that has garnered considerable interest, but it also presents specific challenges, mainly due to the limitations of available resources. In this paper, we focus on identifying and classifying different forms of fake news using unlabeled data, specifically exploring how to use unlabeled data for multi-class classification. The proposed approach categorizes fake news into four forms: satire or fake satirical information, manufacturing, manipulation, and propaganda. Our method employs a relevant approach based on multi-class classification using unlabeled data. The experimental evaluation demonstrates the efficiency of our suggested system.

Strengthening research partner collaboration in higher education for searching innovation through machine learning-based recommender system

Mochamad Nizar Palefi Ma'ady — Thu, 22 Feb 2024 00:00:00 +0000

Academic collaboration is tremendously important for higher education. Multidisciplinary academicians may be grouped as a better research collaboration than the previous one. Therefore, such system is needed, even for a huge number of academicians in a institution. However, existing such recommendation tools are expensive. This paper suggests to develop a system by using machine learning approach in order to search a big academicians data effectively. Hence, with help of standard of Naïve Bayes creates a flexible text search without depending on what select options including research location or case study instead of only research topic. Furthermore, the output of Naïve Bayes, then, is tranformed to percentage display in order to bring ease of understanding the gap of recommendation. It allows the user to choose a possible partner more than one. Therefore, this approach helps reduce time and effort.

Enhancing data security in SAP-enabled healthcare systems with cryptography and digital signatures using blockchain technology

Sonali Shwetapadma Rath, Prabhudev Jagadeesh M P — Thu, 22 Feb 2024 00:00:00 +0000

As the healthcare industry adopts more digital technologies, guaranteeing the security and privacy of sensitive patient data becomes increasingly important. Traditional centralised authentication solutions leave cyber threats and unauthorised access vulnerable. In response, the research demonstrates a novel strategy to improving data security and authentication in a SAP-enabled healthcare system by leveraging encryption and block chain technologies. The research paper discusses the development and integration of a block chain-based decentralised identity management system within the SAP platform. Each healthcare entity, including patients, doctors, and administrators, is given a distinct digital identity that is protected by using base 64 activity through DocuSign protects in SAP platform. The benefits of the proposed solution are assessed using a complete security analysis that measures data confidentiality, integrity, and availability. In comparison to traditional authentication systems, the block chain-based approach is more resilient to cyberattacks and data breaches.

Performance evaluation of deep learning models for detecting deep fakes

Aishwarya Rajeev, Raviraj P — Thu, 22 Feb 2024 00:00:00 +0000

The proliferation of deep fake content in multimedia has necessitated the development of robust detection mechanisms. In this study, a comparative analysis of four state-of-the-art deep learning models for detecting deep fakes is conducted: CNN+RNN, DAFDN, Hybrid Inception ResNet v2, and Xception. The evaluation focuses on their performance metrics, emphasizing accuracy as a primary measure. Through extensive experimentation and evaluation on a comprehensive dataset, the findings reveal notable distinctions among these models. The CNN+RNN architecture demonstrates a commendable accuracy of 94.8%, providing a solid baseline for comparison. Surpassing this, the DAFDN model achieves an accuracy of 97.8%, showcasing superior discriminatory capabilities in identifying manipulated content. Furthermore, the CNN model stands out with an accuracy of 98%, exhibiting remarkable effectiveness in distinguishing between genuine and deep fake media. The comparative analysis delves into the strengths and weaknesses of each model, shedding light on their respective performance levels in detecting sophisticated deep fake content. The observed accuracies underscore the nuanced differences in their architectures and training methodologies, offering insights crucial for selecting appropriate models based on specific detection requirements.

Development of a Solar System for Charging Mobile Phones with customized DC chargers for Rural Areas in Nigeria

Thu, 22 Feb 2024 00:00:00 +0000

The introduction of mobile phones has redefined the world of communication in that it has turned the world into a global village as people can now make contact through phone calls within and across countries at affordable rates. Nowadays, almost every home has a functional mobile phone, be it a conventional, android, or iPhone, among others. These phones use non-self-charging rechargeable batteries that need to be recharged from time to time to meet the demands of the users. However, access to a reliable source of power to meet the energy demand, including mobile phone charging needs, of off-grid rural dwellers remains a global challenge. As a result, this study designed and implemented a solar-powered mobile phone charging system with customized dc chargers for use in remote off-grid areas. The test results showed that the system is very effective for charging mobile phones.