Scandinavian Journal of Information Systems

An Effective Solution Towards Solving the Problem of Deepfake

Anushree Deshmukh and Sunil Wankhade — Mon, 22 May 2023 00:00:00 +0000

The use of advanced AI algorithms is growing daily, resulting in increasingly realistic fake videos with facial superimposition that are being produced by AI, as these videos involve well-known people, their behavior can have severe consequences. The videos have the potential to not only spread false information that harms the status of individuals, businesses, and nations, but also to trigger widespread hysteria and civil unrest. Therefore, to stop the rapid spread of Deepfake it is of extreme importance to identify it first. A novel Residual Network based model that learns inherently distinct patterns that change between a real video and a Deepfake is proposed in this paper as a method for detecting Deepfakes. An Inception Residual Network Version 2 has been trained by the proposed model to successfully differentiate Deepfake videos using different features of a frame-based approach. This work shows that our model can be used to accurately identify Deepfake faces in each video source. This will help security applications effectively reduce the ever-increasing threat of Deepfakes.

Tetrahedral Hyper-Cube: A New Scalable Hybrid Interconnection Network for Massive Parallel Processing

Rashmita Padhi and Nibedita Adhikari — Mon, 22 May 2023 00:00:00 +0000

This paper introduces an enhanced highly scalable, recursive and hierarchical interconnection topology called the Tetrahedral Hyper Cube (THC) for high end computing systems. As compared to the other interconnection networks THC is found to be more attractive in terms of various topological attributes such as node connectivity, edges, average node distance and message traffic density etc. The proposed network is bipartite and highly scalable and also robust in nature than the other popular network topologies like Hyper cube (HC), Folded Hypercube (FHC), Exchanged hyper cube (EC), Mobius cube (MC) and Torcube (TC). With increased node count, the average node distance and message traffic density for THC is relatively small. The main attractive property of THC is its comparatively low link complexity which makes it a superior choice for massive scale computing. The optimal peer to peer minimal routing and broadcasting algorithms for the new network are also presented. The isoefficiency of THC lies between 0.6 to 0.56 for a varied magnitude of dimension. The packing density is four times that of the n dimensional hypercube. The detail analysis reveals that the THC topology is an efficient enhancement of the popular Hypercube topology and its variants.

Robust Malware Detection Leveraging Machine Learning Algorithms

Santosh Tamboli, Dr. Sunil Patekar — Mon, 22 May 2023 00:00:00 +0000

The rampant increase in malware attacks has caused a significant impact on various industries and governments, leading to serious consequences. Malware analysis and detection has become hot topics for research. Malware could be anything that looks malicious or acts like a virus, worm, trojan, spyware, adware, etc. Any suspicious software that may cause harm to the system can be considered malware. Currently, static and dynamic strategies are used for malware analysis and detection, but it is time-consuming and ineffective for identifying malware in real-time. Advanced strategies like machine learning and deep learning are used to determine whether any executable file is malware, and these methods give better accuracy and performance traditional methods. These methods use Portable Executable (PE) file header for malware detection and classification.

Analysis of Probability Model for Example Based Machine Translation

Sapna Thakur and Dr. Priyanka Bhalerao — Mon, 22 May 2023 00:00:00 +0000

Until now, Example-based machine translation (EBMT) systems rely on heuristics to retrieve translation samples. This heuristic of takes time to fix and can make’s algorithm confusing. This article reports the EBMT outcome model. According to the proposed pattern, the system searches for the combination of the instances with the highest probability of the translation instance. The proposed model explicitly manages the EBMTprocess. In addition, standard may contain similar definitions of exemplary definitions. Experimental results show that the proposed model has slightly better interpretation than state-of-the-art EBMT systems.

RTO Management System Using Blockchain

Siyavar, Tejasvi Arora, Lakhan Kumar, Rani Astya — Mon, 22 May 2023 00:00:00 +0000

As the world's population continues to grow, so too does the number of vehicles on the road. It is imperative that drivers possess the proper license and vehicle papers as required by law. However, it can be inconvenient for drivers to carry physical copies of these documents at all times, and the process of verifying them can be time-consuming for traffic police who also need to maintain accurate records of issued challans. To streamline this process, we propose the implementation of a system whereby drivers can register with the RTO services and receive unique login credentials. Through this system, drivers can upload all necessary documents, including their RTO driving license, thereby eliminating the need for hard copies. In addition, drivers can utilize the system to file complaints against any police misconduct.

When traffic police need to verify a vehicle's documents, they can simply enter the vehicle number into the application. The system will then fetch relevant data from the server and display all pertinent information, including emission and insurance details. Moreover, traffic police can use the system to generate challans, which will be linked to the specific vehicle and subsequently paid online by the driver. This approach benefits both drivers and traffic police alike, as it simplifies and automates what can be an arduous process while simultaneously ensuring compliance with all necessary regulations.

Transcriptomics is a rapidly growing field that generates new data that may be used on its own or in combination with existing clinical data for development of new therapeutics, including gene therapy

Moataz Dowaidar — Wed, 01 Mar 2023 00:00:00 +0000

Transcriptomics is a rapidly growing field that generates new data that may be used on its own or in combination with existing clinical data to widen and affect the future of healthcare. While the majority of current applications are limited to research, a growing number of studies suggest that transcriptomics has applications in diagnostics, genomics-driven trial design, and the creation of personalized medicines. Blood samples can be collected in general practice and submitted to a central lab for analysis and interpretation before being provided to the doctor, allowing for greater clinical acceptance of experimental hypotheses. The transcriptome's immense complexity has been revealed by transcriptomics, and we're just beginning to understand how this translates to function, disease, and therapeutic options.

Applying technology to serve tourists without tour guides - A challenge for students learning foreign language

Le Pham Duc Truong — Tue, 23 May 2023 00:00:00 +0000

The tourism industry is one of the most dynamic and rapidly evolving sectors in the global economy. The industry faces significant challenges in providing quality services to tourists, including the need to guide tourists without a guide. With the continuous advancements in technology, there is a growing interest in leveraging technology to address this challenge, particularly for students learning the foreign language. This study aimed to explore the challenges and opportunities of applying technology to guide tourists without a guide. The study focused specifically on foreign language learners and aimed to evaluate the effectiveness of technology in guiding tourists while examining the challenges that may arise during the process. The research methodology involved a comprehensive review of the literature on the use of technology in tourism and language learning. Additionally, interviews were conducted with experts in the tourism industry and language learning. Based on the findings, a prototype application was developed that leverages technology to guide tourists without a guide in Japan. The study revealed that integrating technology in guiding tourists without a guide offers several benefits, including enhanced accessibility, flexibility, and convenience for tourists. Technology also offers the opportunity to provide more personalized experiences that cater to the unique needs of tourists. However, there are challenges that need to be addressed, such as the accuracy of translations and potential technical issues. The study underscores the critical role of technology in the tourism industry, particularly in guiding tourists without a guide. Innovative applications that cater to the needs of tourists can enhance the overall tourist experience and offer valuable language learning opportunities to students. The findings of the study have important implications for the tourism industry, particularly in providing quality services to tourists. The study highlights the need for continuous innovation and the integration of technology to cater to the unique needs of tourists. Furthermore, the study offers insights into the potential benefits and challenges of technology in guiding tourists without a guide and offers recommendations for further research.

Evaluating the Effectiveness of Teaching Japanese Online Courses: CPA-Learning and Coursera

Tran Anh Kiet — Thu, 25 May 2023 00:00:00 +0000

FPT University students are encouraged to use online platforms such as CPA-learning to supplement their self-study and develop language ability in the process of teaching and learning Japanese online. Furthermore, interactive teaching methods such as live chats, Q&A sessions and periodic tests are used to stimulate student participation and active participation, thus ensuring the acquisition of information. The research paper is based on learning through technology with 100 Japanese language learners at FPT University, Ho Chi Minh City. The research paper carried out by quantitative and qualitative methods to determine online learning and communication over a period of one month demonstrates the success of this educational strategy. As a result, students give good comments about the quality of teaching and the development of their Japanese language ability, the quality of teaching and supporting students to achieve their language learning goals.

Use of Mobile App - Detect SOC to digitize and analyse the readings acquiredthrough SOCDK while checking the level of Soil Organic Carbon in agriculture soil from sample farms

Dr. Vinita Gaikwad, Ms. Anamika Dhawan, Ms. Shweta Wahgmare — Thu, 25 May 2023 00:00:00 +0000

The level of atmospheric carbon has increased since the advent of industrial revolution in 18th century. Several innovations and inventions lead to a progressive lifestyle and the consequent burning of fossil fuels. Over a period of time these activities of combustion led to the increase of atmospheric CO2 and eventually global warming. This further disturbed the natural course of seasons across the globe. Today we nearly stand at the threshold of extinction of the globe if the emission of greenhouse gasses is not controlled. Agriculture is one of the prime sectors and is responsible for the contribution of greenhouse gasses, of which CO2 is prominent. As a matter of fact, the Agriculture sector behaves as a source and sink of CO2. Availability of organic carbon in the soil is an indicator that the soil is rich in micronutrients which are essential for a voluminous and good quality crop. Till date several tools of Information Technology have been applied to evaluate the level of SOC in soil. This applied research study probes into the use of Information Technology in the form of a Mobile App - Detect SOC for evaluating the Soil Organic Carbon (SOC). The readings of the Detect SOC App are calibrated against the results of SOCDK kit for the said agriculture soil samples. Based on the evaluated level of SOC the Mobile App provides suggestions to enhance the level of SOC in the sample soil.

Leveraging Educational Data Mining Techniques to Predict Students’ Performance in Campus Placement

Prof. Sonu Gupta, Prof. Rashmi Vipat — Thu, 25 May 2023 00:00:00 +0000

One of the most significant problems in Educational Data Mining (EDM) is predicting students' performance. The objective being identifying students who are at a risk of failure at an early stage so that preventive actions such as continuous mentoring, assisted or personalized learning, continuous assessments can be taken by the educators. In this paper, we have analyzed students’ academic and placement data of past 4 years of our Institute to understand the extent to which academic results affect placement outcomes. We have used Linear Regression, ID3(J48) and Naïve Bayesian data mining models.

Sistemas de Información como Guía Metodológica para la Elaboración y Comunicación de Tesis Innovadoras

Yter Antonio Vallejos Díaz — Thu, 25 May 2023 00:00:00 +0000

The purpose of the study was to evaluate Information Systems (IS) as a methodological guide that allows the elaboration and communication of the thesis of technological innovation in engineering. The research was applied, non-experimental and descriptive transectional, with a sample of 34 professors from the Faculty of Engineering of the National University of Cajamarca, through an 8-item questionnaire. The degree of consistency of the use of the information system as a methodological guide in the preparation and communication of the thesis of technological innovation in engineering was high, on average 80%, with a minimum variance of 7%. The Pearson coefficient of the relationship between the consistency of the elaboration processes and the consistency of the communication processes of the technological innovation thesis was minimal and insignificant (p>0.05) because it is not a linear relationship but a bidirectional spiral. The elaboration and communication of the thesis of technological innovation, was oriented with the information system contained in the proposed methodological guide, in four processes: emanation of the idea, understanding of the product, development of the prototype and establishment of production and marketing criteria.

Study on Machine Learning Algorithms for Reducing Pesticide Spray on Crops

Padma Nilesh Mishra, Shirshendu Maitra — Tue, 23 May 2023 00:00:00 +0000

Machine learning can be used to optimize the application of pesticides in agriculture by predicting the amount of pesticide needed based on various factors such as crop type, growth stage, weather conditions, and soil composition. This can help reduce the amount of pesticide used, increase crop yield, and minimize environmental impact.Machine learning algorithms like logistic regression classification, polynomial regression, and K-nearest neighbor (KNN) can be used to classify which sections of a crop field require repeated pesticide spraying.These algorithms can analyze data on factors such as weather conditions, soil type, pest populations, and previous pesticide applications to determine which areas of the field are most susceptible to pest damage. By using this information, farmers can target their pesticide applications to only the areas that need it, instead of spraying the entire field repeatedly. Thus, in the paper, we discussed accuracy percentages based on different machine learning algorithms.

Detection of DDoS Attack using Machine Learning with Software Defined Network

Owais Farooq, Dr. Mohammad Mazhar Iqbal — Thu, 25 May 2023 00:00:00 +0000

The new systems administration model that decouples the control plane from the sending plane, called programming characterized organizing that eliminates vertical coordination in present inheritance organizations and giving a worldwide perspective on the organization through an organization working framework known as a regulator. This will bring down systems administration costs by software parts that were recently given as equipment, while also enhancing development, network security, administrative quality, and virtualization are commonplace. Notwithstanding the advantages of programming characterized organizing (SDN, for example, an organization broad perspective through the regulator, the regulator is its primary imperfection. More demands to the regulator can be send by the aggressors, than it can process, making it crash and become inaccessible, in this way taking the organization disconnected. The objective of this study is to give a half and half lightweight component that works with the regulator to assist with finding network oddities while utilizing the least assets conceivable. In light of the impediment of assets in SDN, a stream interruption identification framework (IDS) is utilized. In any case, in light of the fact that the stream IDS has a high pace of bogus up-sides, when an assault is identified, the stream is communicated to an entropy mini-computer module in view of the TCP 3-way handshake with an edge indicated. Assuming that the stream is really an assault, a SYN parcel counter forestalls the IP address from flooding the regulator assuming the quantity of bundles is more noteworthy than 50 every second.

Optimizing Energy Efficiency of Wireless Sensor Network through Machine Learning Algorithms

Urooj Sultana, Dinesh Sethi — Thu, 25 May 2023 00:00:00 +0000

Machine learning is a big part of artificial intelligence. Artificial intelligence systems are broad and complex, and they're programmed to solve complicated problems the same way humans would. Machine Learning is an emerging technology that is rapidly increasing in popularity. It is currently being used in a variety of different industries and is popular among many different types of businesses. It’s a field of study where computer systems, software programs, virtual assistants, etc. become capable of automatically learning without explicitly programmed instructions. With Natural Language Processing (NLP), machines can read, understand, and interpret human language. With NLP, machines can perform speech recognition, sentiment analysis, and automatic summarization. NLP and text analytics use machine learning techniques to identify speech parts, entities, sentiment, and other elements of a text using statistical techniques. It is referred to as supervised machine learning when the techniques are expressed as models that are applied to other pieces of text. This paper reviews the concept of Machine Learning, its role in Natural Language Processing, as well as reviews the works in NLP.

Skin Cancer classification: An Advanced Deep Learning Model Architecture for Classifying Dermoscopic Images

Syeda Mahreen, Naresh Sandrugu, Aruna Varanasi — Fri, 02 Jun 2023 00:00:00 +0000

Skin cancer is a widely spread type of cancer that impacts a significant number of individuals globally. As per the data presented by the World Health Organization, almost 2-3 million cases of non-melanoma skin cancers and about 132,000 cases of melanoma skin cancers are identified every year. The survival rate decreases as the skin cancer stage progresses. The timely identification of skin cancer can significantly improve the survival rate of patients by up to 74%. However, it is an expensive and challenging procedure to identify the cancer type in the early stages. Machine learning and deep learning have the potential to make a significant impact on the classification of skin cancer. A robust medical support system for classifying skin cancer from dermoscopic images is required for the prognosis of skin cancer. Therefore the primary objective of our research is to develop a model which automatically classifies skin cancer into distinct types. We conducted several experiments on skin cancer dataset utilizing machine learning and deep learning techniques, including fine-tuning and transfer learning. We observed that the deep learning fine-tuned model, which utilized the Densenet201 architecture to implement fine-tuning, performed exceptionally well across all experiments. Furthermore, we evaluated the performance of all models to classify seven types and further three categories of skin cancer. The accuracy levels achieved by all classification models were notably commendable, meeting the established benchmark for classification.

Framework Models for Theme Park Design: Incorporating Transportability Models and Visitor Preferences

Arati Bachanna & Manoj Patil — Fri, 02 Jun 2023 00:00:00 +0000

In order to evaluate designs in transportation and urban planning, simulating human mobility is often necessary. Our specific focus is on designing and assessing wireless networks and services for new theme park experiences. The effectiveness of certain wireless networks, such as mobile ad hoc networks, is highly influenced by human mobility. To address this, we have created a tool called ParkSim, which allows us to simulate the movement of visitors in a theme park. The tool utilizes an activity-based mobility model that is driven by the visitors' desired activities within the park. Calibration of the tool is based on GPS data collected from an entertainment theme park, and it has been validated against several performance metrics relevant to wireless ad hoc networks. Our plan is to expand ParkSim to evaluate new approaches for balancing the distribution of visitors across different areas of the park.

Analysis of global solar radiation, satellite frequency for the city of juliaca

Fri, 02 Jun 2023 00:00:00 +0000

The path towards a greener economy is leading the world energy scenario to an increasing use of renewable energy sources such as solar, however, in order to make better use of this type of energy, it is necessary to keep the data updated regarding to the amount of solar radiation that falls on each area of our planet. The research aimed to obtain satellite data and compare it with the measurement by the ground station and establish the degree of reliability by the satellite. The mean standard error (RMSE), the mean error (MBE), the determination coefficient (

Smart Cities: Survey On Addressing Challenges In Smart-Cities

Mahesh Kumar Thota*, Prathibhavani P M, Venugopal K R — Fri, 02 Jun 2023 00:00:00 +0000

Technologically modern urban areas that use Information and Communication Technologies (ICT) to increase operational efficiency in sharing information with the public are known as smart cities. The fundamental objectives in smart cities development are sustainable environment, Quality of Life, and cost of living. To harmonious advancement of technology and the quality of citizen’s lifestyles. individuals are moving to smart cities from urban cities. The characteristics of smart cities are Quality of Life (QoL): mainly concentrates on financial emotional wellbeing of citizens, Sustainability: Engages with city governance, climatic change socio economic and citizens health, Smartness: Ensure the improvement of urban city standards, Urbanization: Enhancement of economic growth and interrelation policy formulation and realization. Four main pillars in smart city development are institutional infrastructure to integrate services and city operations with people, physical infrastructure to ensure sustainability, social infrastructure to bring awareness and economic infrastructure to flourish a smart economy. This survey aims to investigate and address the smart-cities challenges by comparing the existing solutions that are compatible with the advanced technological innovations for solving the smart-city challenges. Disaster management, e-Governance, e-transportation, waste management, cyber threats, data management and environmental safety are interconnected to analyze the challenges that arise with the exponential growth of urbanization and population. Technology acts as routine protocol between sustainability and flexibility in development of prospective smart-city infrastructure using Fog-Cloud techniques with Internet-of-Things(IoT) support. Smart city is an application of IoT notion with an unceasing growth of population and urbanization have intensified innovative ways to handle city operations using Fog-computing and cloud storage network support intelligently with minimal human interaction in the generic architecture. Therefore, this paper aims to deliver the essence of smart cities in identifying the optimum technique for providing high data security and resource privacy with less latency and benchmark results. The researchers, those who are working to address the smart-urbanization challenges with technological enhancement to solve the key challenges, act as a real-world implementation mainstream gateway providing limitations and further challenges in smart-city through extensive literature survey.

Determinants of Dividend Policy: An Empirical Study of Non Financial Industries of PSU’s in India

Dr. Suman, Dr. Pooja Gupta, Dr. Rohit Garg, Dr. Abhinav Gupta — Thu, 01 Jun 2023 00:00:00 +0000

This research paper explores the determinants of dividends in case of PSU’s belonging to various industries. The analysis is based on 10-years data covering the period from 2012 to 2021. The review of literature indicates that the major determinants of dividend decision of a firm include primarily, retained earnings, quick ratio, debt to equity ratio, capital expenditure (capex), ROE,EPS, current earnings . The Multiple Linear Regression technique has been used in this study to find out the relationship of determinants of dividend with equity dividend. The study further revealed ROE, EPS, and current earnings are significant determinants and positively related with dividend decision. However, retained earnings and capital expenditure significant but have negative impact on dividend payments.

AI-Guided Energy Optimization In Hvac System

Thu, 01 Jun 2023 00:00:00 +0000

This paper proposes a predictive analysis framework for Heating, Ventilation, and Air Conditioning (HVAC) systems using Reinforcement Learning (RL) techniques. The framework aims to optimize the energy consumption of HVAC systems by predicting the optimal set- points for HVAC controllers. RL algorithms, specifically Q- Learning and Deep Q-Network (DQN), are applied to learn the optimal set-points based on historical data and real-time feedback. The approach is evaluated using real-world data from an HVAC system in a commercial building. The results show that the RL-based approach can achieve significant energy savings while maintaining a comfortable indoor temperature. This study demonstrates the potential of RL techniques for predictive analysis and optimization of HVAC systems using model free deep RL techniques and Influx db as database while analyzing the performance better using Grafana visualization tool.

Risk - Return Analysis of Selected Mutual Fund Companies listed in NSE

Dr R Sathya, Dr Padmaja DV, Dr C Aishwarya, Dr. B. Sivakumar — Thu, 01 Jun 2023 00:00:00 +0000

Since the liberalization process began in 1992, Mutual Funds, which are a component of the Indian Financial Sector, have risen significantly in the economic hierarchy. The capital market reforms and economic expansion that permitted the mutual fund industry's fast rise also provide a clear path for this sector's future development. Only a small portion of savings are now making their way to the stock markets via mutual funds. They are relatively simple, cost-effective, and don't require an investor to choose which securities to buy. Because only 2 to 3 percent of all Indians invest in mutual funds for a variety of reasons, familiarity with the companies that offer these plans is now required. This study will help identify the risk and return of the mutual fund companies that are listed on the National Stock Exchange (NSE).

Parallel mediation analysis of self-image and perceived usefulness between job security, habit, organizational culture and intentionto use AI technologies

Aditi Vasunandan, Sumathi Annamalai* — Thu, 01 Jun 2023 00:00:00 +0000

With technology advancing at a much faster rate today, organisations risk obsoletion if they do not keep up with the fast pace and adoption of artificial intelligence (AI) technologies. The most important aspect of adopting AI technologies is how employees will respond to it. Employees make the core of an organization, and if they do not respond well to changes, the company will be adversely affected. This research report aims at examining whether job insecurity, habit, organizational culture, perceived self-image, and perceived usefulness affects employees’ intention to use AI in the workplace. Data was collected from 204 Indian ITES employees. Three mediation models were hypothesized and explored. For all the three models, parallel mediation analysis was conducted using Hayes’ PROCESS Macro (Model 4). The results from the first model indicate that job insecurity directly affects the intention to use artificial intelligence technology among the respondents. In the second model, both self-image in using technology and perceived usefulness emerged out to be mediators between habit of the respondents and their intention to use AI. From the third model, the results indicate that self-image in using technology and perceived usefulness mediates the relationship between perceived organizational culture and intention to use AI. The study will help the ITES business organizations prepare their employees for radical shifts in technological adoptions to prevent their discomfort. The first step in this process would be to identify the variables affecting their employees’ intention to use artificial intelligence and this research report helps in studying such variables.