Welcome to CSIP 2022

9th International Conference on Signal Processing (CSIP 2022)

December 22 ~ 23, 2022, Sydney, Australia



Accepted Papers
Initial intrusion detection in Advanced Persistent Threats (APT’s)using Machine Learning

Singamaneni Krishnapriya1, T Chithralekha2, 1Research Scholar, Department of Computer Science Pondicherry University, Puducherry, India & Assistant Professor, Department of CSE(CS), Guru Nanak Institutions Technical Campus,Ibrahimpatnam, India, 2Professor, Department of Computer Science, Pondicherry University, Puducherry, India

ABSTRACT

Cyber assaults have resulted in significant financial losses and the disruption of essential governmental services in recent years. Among these cyber-attacks, theAdvanced Persistent Threat (APT) has become a major security issue to cyber security,affecting a number of businesses and organizations. The APT attacks are targeted andhave a specific objective behind the attacks. An attacker builds an Entry point to the target network as part of the APT attack. This is commonly accomplished by infectingthe targeted system with malware that leaves a backdoor open for future access. The spear phishing email is a prevalent method for breaking into a network for establishingthe initial foothold into a target network. The phishing mails may contain malware attachment or Link which leads to website injected with malware code. This research offered an authentication-based detection approach for spear-phishing emails. The pro-posed technique takes the trust email dataset’s stylometric, gender, and personality traits and extracts them.

KEYWORDS

Advanced Persistent Threats, Cyber Security, Gender feature, Spear-phishing, Personality features, Personality features, Stylometric feature.


Impact of Digitalization on Shaping Consumer-Centered Smart Healthcare System - A Comprehensive Study

Bibhu Dash, School of Computer and Information Sciences, University of the Cumberlands, KY USA

ABSTRACT

We are on the verge of the Fourth Industrial revolution, in which digitalization, data analytics, and insights play a critical role in our daily lives. The COVID-19 outbreak has created unprecedented uncertainty for many firms. Healthcare organizations across the care continuum were confronted with new difficulties, forcing them to learn swiftly, adapt, and reinvent the way they offer traditional care. As a result, the notion of cost-effective, anytime, and anyplace care has entered the mainstream, forever altering the rate at which we obtain healthcare. As a result, healthcare industries across the globe started to rethink their tactics to be ready for the future and this new breed of customer demand. With rapid technological evolution, this paper analyzes the trends and recent shifts in the healthcare industry and how it is trying to be more customer-responsive by using emerging technologies.

KEYWORDS

Smart Healthcare, Digitalization, EHR, CCH, eHealth, mHealth, Cyber risk, IMT.


Trackid – A Location Tracking Application for Children on School Transport

Rachel Karanja and Esther Khakata, School of Computing and Engineering Sciences, Strathmore University, Nairobi, Kenya

ABSTRACT

Education does not only entail children sitting in class to read books and study, but it also involves activities outside the classroom such as co-curricular activities, transport system, religious activities, and cleaning but unfortunately, they are not given as much thought as anticipated. In the school transport sector specifically, there has been a growing concern among parents on the safety of their children when they are on their way back home to school. Some parents are not able to take their children to and from school due to work demands and frequent traffic snarl-ups, therefore they choose to enroll their child into the school transportation system to cushion them of that burn. The problem however is that even with the school transportation, parents are not able to track the movements of their children while they are on transit to and from school and this has brought about many cases of children getting lost and not arriving home and this results into a blame-game between the school and the parents. The school, in this case, the drivers and the teachers are unable to account for the children and this brings further mistrust, and hence the parents are forced to either bear with the situation or pull out their children from the program altogether and look for alternative options. This paper proposes a system that will solve this issue by presenting a web application that will enable the school, parents and the drivers be accountable for the children. The parents can be constantly updated on the location of their children while they are on transit and provide the school with a way of tracking students that use the school transportation system. This will allow all the stakeholders to communicate efficiently and in turn ensure children’s safety.

KEYWORDS

Tracking, Location Tracking, School Children, School Transport.


A Comprehensive Overview of Artificial Intelligence

Kouassi Konan Jean-Claude, Faculty of Computer Science via distance learning, Bircham International University, Madrid, Spain

ABSTRACT

Nowadays, we remark that breakthroughs in the field of AI suggesting its similarity with human beings, tremendous diversity of subfields and terminologies implied in the AI discipline, huge diversity of AI techniques, mistakes of AI and hype could lead to confusion about a clear understanding of the field. In some cases, misunderstanding about AI led to hype, firing, and rude criticism even among many senior experts of the AI domain. In this paper, we proposed a “Comprehensive Overview of Artificial Intelligence (AI)” so that everyone (starting by newbies), be able via clear insights, to rapidly make the difference. In our dissertation we also showed how the AI realm is held by well-known Theories of Intelligence and related AI Concepts that perfectly match the current technological advances in the field. And of course, we did not forget to provide a clear insight of Ethical concerns about Artificial Intelligence.

KEYWORDS

Artificial Intelligence, Machine Learning, Weak AI, Narrow AI, Artificial General Intelligence, Strong AI, Superintelligence, Machine intelligence, Theories of Intelligence, Ethics and Risks of AI.


Cryptanalysis of the SHA-256 Hash Function

Olga Manankova and Mubarak Yakubova, Almaty University of Power Engineering and Telecommunications name after Gumarbek Daukeev, Almaty, Kazakhstan.

ABSTRACT

The research of the strength of a hashed message is of great importance in modern authentication systems. The hashing process is inextricably linked with the password system, since passwords are usually stored in the system not in clear text, but as hashes. The SHA-256 hash function was chosen to model the attack with arc tables. An algorithm for constructing a rainbow table for the SHA-256 hash function in the Java language is proposed. The conditions under which the use of rainbow tables will be effective are determined. This article aims to practically show the process of generating a password and rainbow tables to organize an attack on the SHA-256 hash function. As research shows, rainbow tables can reveal a three-character password in 3 seconds. As the password bit increases, the decryption time increases in direct proportion.

KEYWORDS

Hash function, cryptanalysis, Rainbow tables, SHA-256, Java.


Tomatoes Septoria Leaf Spot Disease Detection based on Artificial Intelligence in North Province of Rwanda

IRAGIRE Viateur1 and KOMEZUSENGE Joseph2, 1Department of Electrical and Electronics Engineering, Rwanda Polytechnic, Rwanda, 2Software Developer, Kigali-Rwamagana, kigali, Rwanda

ABSTRACT

Safe plants are a key factor in having a significant harvest. In this regard, the quantity and quality of the products from agriculture is found by monitoring the life of plants from the initial state till the final stage. We found that the quantity and quality of tomatoes in Rwanda especially in the north province are affected by Septoria Leaf Spot Disease. In this paper we did research on technology for detecting and monitoring tomato Septoria Leaf Spot Disease. Large- or smallscale fields of tomato need systems that are able to detect plant leaf diseases as many farmers experience significant difficulties in controlling and monitoring the daily growth of their tomato. Traditionally, it requires farmers to go to the farm and observe the state of tomato plant leaves even at remote distances or areas.

In this study, a useful Tomatoes Septoria Leaf Spot Disease Detection based on Artificial Intelligence in agriculture of tomatoes was developed. This system uses database that stores pictures of tomato leaf contain symptoms of Septoria Leaf Spot Disease and Python language for prediction, in that prediction a system compare image of tomato leaf taken by digital camera located in farm and the stored image of tomato leaf contain symptoms of Septoria Leaf Spot Disease and image processing algorithm design, then the predicted answer from the system send it to farmer using microcontroller to control electronics circuit and GSM which is used to send SMS to notify farmers the status of their tomatoes in remote area and python used to predict whether the received photos of tomato leaf from camera in farm was affected by septoria leaf spot diseases.

This research was achieved by presenting both circuits with electronic components and flow charts showing the process and implementation of a prototype which is linked to python. The system help farmers to reduces a large work of monitoring diseases in a big farm of tomatoes and the reduction of economic losses particularly in agriculture sector when tomatoes were affected by Septoria leaf disease ,it also used to detect different disease affect tomatoes ,from the result of our system will help farmer to control their tomatoes and given then first response for making decision of cure the affected tomatoes.

KEYWORDS

Tomato Leaf, Septoria leaf spot disease, Artificial Intelligence, python, GSM, Microcontroller.


Integration of Machine Learning in Agile Supply Chain Management

Vivek Ghabak and A. Seetharaman

ABSTRACT

One essential component of any manufacturing industrys success is a successful supply chain. Customer expectations have increased due to the quick growth of information and communication technologies, which has also made the world more competitive. Global production and manufacturing have now transitioned to Industry 4.0. The "Internet of Things," "Big Data," and "Artificial Intelligence" are the dominant digital technologies in this. Consequently, in the near future, supply chain management (SCM) will manage not only the flow of raw materials, semi-finished goods, finished goods, and services from the manufacturer to the customer, but also the flow of the most recent data for current and future supply chain visibility & sustainability. For this many companies around the world have tapped one of the advanced technique - Machin Learning (ML) for early risk identification & management, material planning & forecasting, raw material price forecasting etc. As a result, purpose of this paper is to explain a conceptual framework which illuminates factors influencing integration of Machin learning (ML) in Agile Supply chain management, its benefits, and challenges to implement.

KEYWORDS

Machine learning, Agile supply chain management, Industry 4.0, Supply chain risk management, advanced technologies.


Integrated Framework for Understanding the Impact of Machine Learning and artificial Intelligence to Improve Customer Service in Healthcare Industry

Debopam Raha1 and Dr. A Seetharaman2, 1Doctor of Business Administration, SP Jain School of Global Management, India, 2A Seetharaman, Professor and Dean, SP Jain School of Global Management, Singapore

ABSTRACT

This paper studies how artificial intelligence (AI) and machine learning (ML) have an impact on healthcare service delivery, specifically on patient care in the Indian healthcare scenario. The study tries to understand the impact in different areas of health service, both the utopian and dystopian views, and the antecedents of AI in healthcare services. It then exploresthe Indian healthcare scenario to understand the applicability of AI there. The paper performs a detailed literature review on AI in healthcare by looking into the journals and articles published in last two years. A conceptual framework was developed based on the findings and the gaps identified in the literature. Five independent variables were identified: data governance, workforce competency, patient voice, predictive medicine, security and privacy. Five sub variablesunder each of the dependent variables were further identified and a conceptual framework was developed to measure patient experience. This work provides a novel framework integrating different factors while discussing several barriers and benefits of AI-ML based health. In addition, five insightful propositions emerged as a result of the main findings. Further quantitative study can be done to establish the relationship between the factors and establish the validity of the model.

KEYWORDS

Artificial Intelligence,Machine Learning, Healthcare, Data Governance, Workforce Competency, Patient Voices, Predictive Medicine.


Active Learning Entropy Sampling based Clustering Optimization Method for Electricity Data

Wang Qingnan and Zhang Zhaogong, School of Computer Science and Technology, Heilongjiang University, HeiLongJiang, China

ABSTRACT

Clustering is a crucial part in the field of data mining, and common clustering methods include division- based methods, hierarchy-based methods, density-based methods, and grid-based methods. In order toimprove the accuracy of clustering, an optimization study is made mainly for the division-based methodFCM clustering, and an FCM clustering method that integrates active learning and principal component analysis (PCA) is proposed. The method first uses principal component analysis to reduce the dimensionality of the data to reduce the computation of electricity data, then trains the sample model by active learning, and introduces the entropy (Entropy) method in the uncertainty sampling method, the larger the entropy means the greater the uncertainty of the sample, and the smaller the entropy means thesmaller the uncertainty of the sample, so as to filter the electricity data, and finally the electricity data are clustered by FCM clustering The power data is finally categorized by FCM clustering, and with the proliferation of power data, the power data can be more accurately categorized using this method toachieve the stability of the power grid as well as the utilization rate. Experimental results on threedatasets show that this method improves the accuracy of power data clustering by up to 2 percentagepoints compared to the traditional clustering method without active learning, and achieves good resultsin each dataset compared to other methods.

KEYWORDS

Active Learning, Data Mining, FCM Clustering, Principal Component Analysis, Unsupervised Learning.


Segnn4Slp:Structure Enhanced Graph Neural Networks for Service Link Prediction

Zekun lu, Qiancheng Yu, Xia Li, Yufan Yang and Chen Tang, Department of Computer Science and Engineering, North Minzu University, Yinchuan City, Chnia

ABSTRACT

Learning network representations of Web services plays a critical role in the service ecosystem and facilitates many downstream tasks, e.g., service composition, service recommendation, service clustering, and service classification, etc. However, the performance of most of the existing approaches is limited by the sparse and non-interaction relationships between services. Considering these shortcomings, by proposing Structure Enhanced Graph neural network for Service link prediction. SEGNN4SLP introduces the path labeling method to capture surrounding topological information of target nodes and then incorporates the structure into an GAT model. By jointly training the structure encoder and deep GNN model, SEGNN4SLP fuses topological structures and node features to take full advantage of graph information. To evaluate the proposed method, extensive experiments are conducted on a real API dataset and the results show that the proposed method outperforms the state-of-the-art methods in API link prediction.

KEYWORDS

Network Representation, Web Service, Mobile Network, Graph Attention network, Link Prediction.