Skip to main content
Top

2021 | Book

Proceedings of Fifth International Congress on Information and Communication Technology

ICICT 2020, London, Volume 2

Editors: Prof. Dr. Xin-She Yang, Prof. Simon Sherratt, Dr. Nilanjan Dey, Amit Joshi

Publisher: Springer Singapore

Book Series : Advances in Intelligent Systems and Computing

insite
SEARCH

About this book

This book gathers selected high-quality research papers presented at the Fifth International Congress on Information and Communication Technology, held at Brunel University, London, on February 20–21, 2020. It discusses emerging topics pertaining to information and communication technology (ICT) for managerial applications, e-governance, e-agriculture, e-education and computing technologies, the Internet of Things (IoT) and e-mining. Written by respected experts and researchers working on ICT, the book offers a valuable asset for young researchers involved in advanced studies.

Table of Contents

Frontmatter
Novel Methods Based on CNN for Improved Bacteria Classification

Recent times have witnessed extensive use of deep learning in both supervised and unsupervised learning problems. One of these models is convolution neural networks (CNN) which has outperformed all others for object recognition and object classification task. Although these convolution neural networks have achieved exceptional accuracies, still a huge amount of iterations create chances of getting stuck in local optima makes it computationally expensive to train. To handle this issue, we have developed some hybrid methods using, particle swarm optimization (PSO), genetic algorithm, and autoencoders for training CNN. We have taken the images of bacteria and classified them into four different genera (E. coli, Listeria, Salmonella, and Staphylococcus) to measure the performances of these models. Through this study, we concluded that evolutionary techniques can be used to train CNN more efficiently.

Chahes Chopra, Rohit Verma
Plasma Temperature Classification for Cancer Treatment Based on Hadron Therapy

A plasma is an ionized gas that consists of free electrons, ions, and molecules, or atoms. Plasmas are characterized by many characteristics, including temperature, ionization intensity, and density, where plasmas can also be graded in a number of ways by their severity and estimations of the type they represent. Here, we classified four types of plasma based on the temperature; Hot Plasma (Thermal Plasma), Warm Plasma, Cold Plasma (Non-thermal Plasma), and ultra-cold plasma in the presence of three ions; Carbon, Neon, and Oxygen, respectively. Then, we consider the limiting factors of laser-plasma acceleration based on those four plasma types. By using laser-plasma acceleration, however, Hadron therapy reflects the significant contribution to patient care and many exciting developments. Hadron treatment, in a wide range of cancers, can generate enduring loco-regional disease control. The laser-plasma acceleration is more advantageous due to compact size compared to conventional accelerators of charged particles used in the Hadron therapy centers of today. In this paper, we demonstrate that oxygen ion in attendance of ultra-cold plasma leads to better cancer treatment by increasing the delivered energy to the tumor.

Vala Mehryar Alviri, Sheida Asad Soleimani, Morteza Modarresi Asem
On the Specific Role of Electronic Document and Record Management Systems in Enterprise Integration

The paper aims at providing practical insight into the specific role of electronic document and record management systems (EDRMSs) in the process of enterprise integration (EI) to provide organizational data flow. Based on a case study with the ERDMS Amphora, we identify enabling factors for enterprise integration in order to provide interoperability, where we consider both technological and organizational factors. An EDRMS is not a standalone system, and it must interact with many systems used in organization. In this paper, we discuss some certain information flows based on the EDRMS Amphora and related system.

Ingrid Pappel, Sidra Butt, Ingmar Pappel, Dirk Draheim
Robotic Tool to Multi Time-Lapse Videos Creation

There is scientific research in several areas of knowledge that make use of the time-lapse technique. Tidal, rocks, and glaciers movements, which take up to months to occur, can be seen in a few seconds with this technique. Swamps, deserts, and even submerged areas are places of difficult access and need to be observed. There is always a high cost involved in the process of capturing time-lapse images. In this paper, a robotic tool is presented that meets the operational requirements to support these tasks. Among these new features is the ability to use idle time in the process of capturing images for this type of media. Its built-in camera is moved in different directions, being able to capture images for the generation of multiple time-lapse videos, using only one device. Because of this, this robotic tool is a facilitator to the construction of scientific, entertainment, and educational projects that make use of this type of unconventional data.

João Roberto de Toledo Quadros, Roberto de Castro Souza Pinto
Accomplishment of Multimedia Ontology for the Tunisian Archaeology Field

Interoperability is a key concept in what is known today as “data web” because it is by releasing the data from the application constraints that it is possible to easily navigate and react to additional information that complements the initial resource. However, while being released from the web applications, the information loses its contextualization and it will then be essential to document it by the addition of metadata. To cover the semantic richness of an object and the diversity of the information attached to it, there are computer ontologies designed to provide more meaning to the beacons and contextualize more information about a domain. Nevertheless, until now, we have not found any ontology for the Tunisian archaeological field. In fact, we aim to develop a multimedia ontology for the Tunisian archaeology domain, in order to structure the diversified knowledge involved in this field. The development of ontology will be realized in six steps which are described in this paper.

Degachi Haifa, Yengui Ameni, Neji Mahmoud
Ontology-Based Information Extraction Framework for Academic Knowledge Repository

Extracting required information from the huge academic repository in an efficient way is a tedious job as the repository grows day by day. Getting user satisfaction in searching is the primary concern always. Academic search engines are mostly used by research scholars, scientists, faculty, and students for their academic purposes. The use of ontology in information extraction helps an academic search engine to perform its operation effectively. As the research activities are going on in an extensive pace in all domains, new specialized areas are evolved. This paper proposes a novel framework that incorporates a dynamically evolving ontology that focuses on expanding the existing Computer Science Ontology based on the new findings in existing areas. The Word2Vec model helps to identify new keywords for the expansion of the ontology. The keywords given by the Word2Vec should satisfy a confidence score above a threshold value and only these keywords will be used for updating the Computer Science Ontology. The dynamic semantic web helps a user to retrieve the specific document, and thus provides a high degree of personalization experience.

Veena Jose, V. P. Jagathy Raj, Shine K. George
PLS Approach to Measure the Impact of Digital Transformation on User Satisfaction: Case of Local Public Services in the Northern Region of Morocco

This article means to propose a theoretical framework model to quantify the effect of digital transformation on the satisfaction of users of public administrations. In this way, we proceed with a hypothetic-deductive reasoning. A survey was carried out among 110 users of the public services, and the evaluation of our model was carried out using the partial least squares (PLS) structural equations approach via the software SPSS 21 and XL-stat (version 2018) for exploratory factor analysis and verification of reliability and validity. The results of our study show that digital transformation of the public administration positively affects the satisfaction of its users.

Fadoua Laghzaoui, Najoua El Abbas El Ghaleb, Mohcine Ben Sabih El Imrany
SB-Spectral Clustering Algorithm for Clustering and Validating Sensor

Due to the use of WSN technologies these days, there have been many developments in WSN. And many types of researches have been proposed for the sensor nodes validation in the spots. The efficiency of the WSN is strongly relying on the lifetime of the network. Validation is the most common approach used in WSN to increase the lifetime, the efficiency and the performance of the wireless network. In this paper, a validation and clustering mechanisms are presented by using SB-SPC algorithm for specified the bad sensor nodes in the applications and excluding them from the application to prolong the functioning of WSN. SPC algorithm determines the most accurate not working sensors with the help of the eigenvalues and eigenvectors by using receive signal strength (RSS). The result of the SB-SPC algorithm shows that this approach specifies the almost not working sensors properly, deleting them from the domain, and achieve good performance in the wireless network.

Abdo M. Almajidi, V. P. Pawar
Analysis of Precoding-Based Pilot Decontamination Methods in Massive MIMO Systems

During transmission of information, massive multi-input multiple-output (MIMO) antenna system encounters inter-cell interference, called pilot contamination that limits the capacity. Recent developments in massive MIMO technology have provided a plethora of methods to reduce pilot contamination using precoding in massive MIMO. However, a trade-off exists between the system performance based on achieved spectral efficiency in relation to the sumrates and bit error rates and complexity in terms of iterations count, the number of optimization variables and the relative time consumed to resolve the optimization problems. The current literature contains insufficient information to address this trade-off, a consequence that hinders the advancement of research in pilot contamination. This study covers the gap through a detailed review of various linear and nonlinear schemes centered on the two contending metrics, namely spectral efficiency and complexity. A systematic review approach is adopted to analyze different related studies and their associated methodologies, results and limitations. Moreover, we provide recommendations for future research.

Angelina Misso, Baraka Maiseli, Mussa Kissaka
A Quantitative Assessment of the Television White Space in Tanzania: A Case of Coastal and Morogoro Regions

In this paper, we establish suitable propagation models and use them to estimate the potential of TVWS for Coastal and Morogoro regions in Tanzania. Four mostly used propagation models, Free Space, Egli, Ericsson and Hata were analyzed. We established an AM-Ericsson model from the Ericsson model, which was closest to the measured results in rural areas but still with a 25 dB difference. The results showed that Hata was the suitable model for semi-urban areas and the established AM-Ericsson model had more than 20 dB improvement from the Ericsson model for rural areas. The study found that there is up to more than 95% TV channels availability for Coastal region and 97% availability for Morogoro region. This research provides a baseline information to facilitate the estimation of the potential TV channels for secondary use in Tanzania. The study also guides a proper regulation and planning of spectrum in the country.

Moses Ismail, Mussa Kissaka
Potentials of Digital Business Models in Tourism—Qualitative Study with German Experts

Digitization is one of the most discussed topics in recent years. The traditional business models are also massively affected by digitization. Still, the impact of digitization and the potentials for business is only being analyzed in detail by few. Therefore, this research work focuses on the potential of digital business models in the tourism industry. The results are based on empirical data from German experts. The data were collected and analyzed using grounded theory. The effects of digitization, such as new technologies or changed customer needs, change the usual business models. In the future, many business models will concentrate on digital technologies. In particular, rating platforms or sharing economy platforms will play an even greater role in tourism. In addition, social networks such as Twitter, Facebook or Instagram will strongly influence the tourism industry. These new technologies will change and shape the tourism industry in the future. More and more offers are available for customers who enjoy traveling. The aim of this qualitative study is to identify the potential of new digital business models in the respective markets.

Ralf Härting, Raphael Kaim, Lisa Fleischer, Alexander Sprengel
An Efficient and Secure User Authentication Protocol for Smart Card Users

There are so many communication schemes available today for the communication between authorized remote users and servers over an insecure network. These types of authentication schemes generally use the password for the authentication. To have the communication between remote users and servers, many researchers have proposed remote user authentication techniques using the smart card. The advantage of using the smart card is the storage availability and the computation speed. In many proposed protocols, authors argued that their protocol is secure and efficient against any type of attack. Unfortunately, many model fails against the off-line password guessing attack, and with these schemes, wrong password detection is not easy. Also, as many schemes use RSA cryptosystem to offer the authentication, it adds computational overhead to system which is not suitable for the smart card applications. In this paper, efficient smart card-based authentication protocol using elliptic curve cryptography (ECC) is proposed. This proposed scheme has faster computation as compared to the available schemes, and it is secure against variable attacks.

Javed R. Shaikh
Composite Routing Approach for Vehicular Delay-Tolerant Networks

In the vehicular delay-tolerant networks, (VDTN) association between the sources to the destination is not always achievable at any required period. Consequently, the carrier node saves the message in its intrinsic buffer until an opportunity occurs for forwarding. Fix nodes facilitate in message storage and message relaying. It also helps in improving the performance of VDTN. Considering the mobility of nodes, the bit error rate is high whereas the bit error rate in fixed nodes is comparatively low. In VDTN, bit error rate is not considered in most of the routing schemes. In this article, a composite routing approach is introduced to conquer aforementioned issues. Some features of vehicular ad hoc networks (VANET) are associated to PRoPHET routing protocol for VDTN. The propagation models of VANET are executed for mobile node communication and without it for VDTN. The effect of environmental hindrance is also considered, and this can be either positive or negative. This makes the composite routing approach two-dimensional and much competent. The simulation and performance analysis of the composite approach is done via opportunistic network environment (ONE) simulator. Results show that the composite routing approach outperforms the PRoPHET wrt. delivery ratio and average delivery delay.

Sudhakar Pandey, Sanjay Kumar, Vishakha Chourasia
Emergency Detection with Environment Sound Using Deep Convolutional Neural Networks

In this paper, we propose a generic emergency detection system using only the sound produced in the environment. For this task, we employ multiple audio feature extraction techniques like the mel-frequency cepstral coefficients, gammatone frequency cepstral coefficients, constant Q-transform and chromagram. After feature extraction, a deep convolutional neural network (CNN) is used to classify an audio signal as a potential emergency situation or not. The entire model is based on our previous work that sets the new state of the art in the environment sound classification (ESC) task (Our paper is under review in the IEEE/ACM Transactions on Audio, Speech and Language Processing and also available here https://arxiv.org/abs/1908.11219 .) We combine the benchmark ESC datasets: UrbanSound8K and ESC-50 (ESC-10 is a subset of ESC-50) and reduce the problem to a binary classification problem. This is done by aggregating sound classes such as sirens, fire crackling, glass breaking, gunshot as the emergency class and others as normal. Even though there are only two classes to distinguish, they are highly imbalanced. To overcome this difficulty, we introduce class weights in calculating the loss while training the model. Our model is able to achieve $$99.56\%$$ 99.56 % emergency detection accuracy.

Jivitesh Sharma, Ole-Christoffer Granmo, Morten Goodwin
The Application of Finite Element Method to Analyze Passive Failure and Deformation Mechanisms of Soil in Front of Tunnel Face

In the last many decades, replying to the urgent needs for the infrastructure construction, shield tunneling has been conducted widely in urban areas. Although the technique has been advanced recently owing to the accumulation of practical experiences and the progress of mechanization, there still remains many unknown problems especially in soft ground tunneling. For tunneling, one must always consider not only the stability of a tunnel itself but also the surface settlement due to deformations of soil around the tunnel. Although the patterns of the surface settlements differ for different soil conditions and methods of tunneling, many field observations and model tests show that the troughs of surface settlement can be approximated by the error functions or Gaussian normal distribution curves. This article concentrates on analyzing passive failure and deformation mechanisms of the soil in front of tunnel face due to tunneling.

Nguyen Anh Tuan
The Working Capacity of Vietnamese Local Civil Servants

Local civil servants are those who directly deal with the people’s requests, guaranteeing their lawful rights and interests in accordance with the law. In each country, they play a really crucial role in the implementation of state policies and political goals at the grassroots level. The performance of governmental agencies heavily depends on the working capacity of the contingent of local civil servants, which is confirmed by many researchers. Therefore, this research focuses on analyzing the working capacity of Vietnamese local civil servants through such contents as the occupational practice capacity, the Sense and Responsibility for the work, the Attitude to serve the people, and the Additional capacity. On the basis of analyzing the criteria of the civil servant’s capacity, the authors designed the questionnaire and conducted a survey of people’s opinions on the working capacity of civil servants in governmental agencies representing all three provinces of Vietnam which are Hai Phong, Hue, and Can Tho. The survey was conducted selectively, in which the people answering the questions were the residents who had contacted and worked with civil servants at local government agencies more than 5 times. The research results show that the people assessed the occupational practicing capacity as the factor the most powerfully affecting the working capacity of civil servants, followed by the Attitude to serve the people and the Sense and Responsibility for the work, respectively, and finally is Additional capacity. From the above research results, the authors propose a number of suitable solutions to improve the quality of local civil servants in Vietnam. The solutions include: (1) Strengthening professional retraining for civil servants; (2) Regularly raising awareness of working sense and responsibilities and service attitude of civil servants; (3) Providing civil servants with basic knowledge and skills in foreign languages and information technology to well perform professional tasks.

Ngo Sy Trung, Do Huu Hai, Vu Thi Yen Nga, Tran Thi Hanh
AI-HI…The Technological Transgenders (AI-HI Techno-Trans)

The evolution of technology (specifically artificial intelligence and machine learning) in the recent time has made the human life easier, fast, and providing real-time solutions, but at the same time, its negative impacts are also simultaneously affecting the human intelligence and human behavior. This paper discusses some observations on the impact of technology (AI) upon the human intelligence (HI) and human behaviors, and the possible threats as an outcome of the uncontrolled and excess use of the technology enabling loss in the human intelligence, resulting in the emergence of ‘AI-HI…The Technological Transgenders’ abbreviated as ‘AI-HI Techno-Trans.’ The paper defines Techno-Trans Children, Techno-Trans Youth, and Techno-Trans Robots, who would collectively forms a Techno-Trans Society enabling the benefits and threats together. The rapid increase in the development of artificial intelligence (AI) technology, along with the simultaneous decline in human intelligence (HI) might be the outcome of the excess use of the technology, and dependency upon it at large toward the formation of the Techno-Trans Society.

Vijay Singh Rathore
Reliability of the Kinovea bidimensional System for Measuring Buccal Movement in a Sample of Italian Adults

The purpose of this study was to assess reliability of Kinovea bidimensional system for measuring buccal movement in frontal or lateral perspectives. This software was used to analyse participants’ buccal movements in the anticipatory phase of swallowing. The participants assumed five different types of stimuli diversified on the basis of the volume of the bolus and the size and type of aid (straw or spoon). Thirty-two measurements based on movements of the mouth were recorded using the Kinovea programme. Reliability of 32 measurements was evaluated with Cronbach’s alpha and Pearson’s correlation coefficient. The results of this study show that the scale has an optimal internal consistency and correlation between items. Thus, using Kinovea software to analyse and study buccal movement can provide clinicians and researchers with important information about the anticipatory phase of swallowing.

Giulia Rossi, Giovanni Galeoto, Antonio Amitrano, Elisa Leonetti, Livia Flammini, Marco Tofani, Anna Berardi
A Technique to Analyze a Cassandra Tombstone for Identification of Operations

High scalability, flexible storage of data, linear performance, and many other strong features made Cassandra database a popular database. The Cassandra is one of the column-based NoSQL databases. Cassandra performs faster write operations with an additional overhead of tombstone. Whenever logical delete operation is performed, it will result in a tombstone. This tombstone value is useful to update the other replica about the deletion. The major problem with tombstone is that it is not only generated with delete operation but there are other factors also which generate this tombstone. This paper is an effort to identify the key of deleted data and other reasons for tombstone generation. The technique is proposed for the same. This paper is the first attempt to work on Cassandra’s tombstone.

Vinod Pachghare, Rupali Chopade
Computer-Assisted Self-Training for Kyudo Posture Rectification Using Computer Vision Methods

To some individuals, particularly archery students, perfecting the art of Kyudo is of utmost importance. These devoted students are always trying to correct their posture because it plays a significant role in effectively shooting at the target. However, due to the lack of attention from instructors, students are often forced to train on their own without any guidance. It is difficult for students to analyze their own faults because the shoulders, hips, and feet should be in line with another, parallel to the floor and straight to the target. The proposed solution is, therefore, a system that aims to assist students in correcting their posture. The system will classify the technique presented by the user and using PoseNet, the system will output coordinates and draw a skeleton structure of the user’s technique along with the instructor’s technique. The coordinates will then be measured for similarity and appropriate feedback is provided to the user. The results for classification, using CNN and SVM showed an accuracy of 81.25% and 80.2%, respectively. The results indicate the feasibility of the approach, however, improvement is required in certain areas. Recommendations for improving the approach are discussed.

Wardah Farrukh, Dustin van der Haar
Accident Detection Using Time-Distributed Model in Videos

CCTV surveillance cameras are installed in the majority of roads and highways these days; therefore, it generates millions and millions of hours of data, thus captures a variety of real-world incidents. Road accidents are one of the most severe and fatal incidents, which disrupt the smooth flow of traffic as well leading to wastage of time and resources. Detection of accidents not only helps us to save the life of victims, but also helps in reducing traffic congestion. In this, we have proposed a framework for accident detection based on hierarchical recurrent neural network. The framework localizes and identifies the presence of road accidents in the captured video. The framework contains a time-distributed model which learns both the temporal and spatial features of the video, making the framework more efficient. The proposed approach is evaluated on the dataset, built by obtaining recorded road accident videos from YouTube. Results demonstrate the applicability of our approach performs, accident detection, and localization effectively.

Garima Gupta, Ritwik Singh, Ashish Singh Patel, Muneendra Ojha
Analysis of Traditional Computer Vision Techniques Used for Hemp Leaf Water Stress Detection and Classification

Cannabis sativa L. has risen in popularity due to its large variety of uses and environmentally friendly impact. C. sativa L. is extremely sensitive and displays phenotypic responses to water stress in its leaf and stem structure. Optimizing the use of water in the agricultural process of cultivating hemp requires the determining of the water potential in the hemp plant. Computer Vision techniques to determine water potential can be used as opposed to traditional destructive and complex to implement techniques. The goal of this study is to prove that water stress detection in hemp leaves can be achieved using computer vision as well to create a model and compare computer vision techniques. This study used a dataset pooling technique to create the dataset of hemp leaves. The dataset is split randomly at an 80–20% ratio of training data and testing data, respectively. Two derivatives of the traditional pattern recognition pipelining model were used. The first pipeline employed traditional computer vision techniques such as Canny Edge Detection, Contour Analysis, SIFT, and SVM Classification. The second pipeline embraced an object detection approach by implementing Haar Cascades. The results of the study vary greatly leading to researchers to believe that more work needs to be done to improve performance.

Waseem Shaikjee, Dustin van der Haar
Active Surface Calculation Method Using Kinetic-Hydrodynamic Model

The problem of flow over a thin plate of infinite magnitude installed across the flow is considered. The frontal surface of the plate absorbs gas. For calculations, we used a mathematical flow model containing a combination of the Navier-Stokes-Fourier model and the model kinetic equation of polyatomic gases. The features of the mathematical implementation of the combined model are described. The calculations were performed for a supersonic flow with a Mach number of 2.31 for a Knudsen number of 0.1…0.001 and a plate surface absorption coefficient from 0 to 1. The obtained flow fields were compared with solutions of the model kinetic equation of polyatomic gases. The drag coefficient of the plate was compared with known experimental data. For all considered flow parameters, a satisfactory agreement with the known data is obtained. It is shown that there are no gaps in the derivatives of gas-dynamic parameters in the joining region of the kinetic and hydrodynamic components of the model. The increase in the computational efficiency of the model with respect to the solutions of model kinetic equations is estimated. The conclusion is drawn on the suitability of the considered kinetic-hydrodynamic model for describing highly nonequilibrium flows.

Yury Nikitchenko, Sergey Popov, Alena Tikhonovets
A Three-Layer Architecture for Intelligent Intrusion Detection Using Deep Learning

Recently, the increasing number of machine learning algorithms has been used in network intrusion detection system (NIDS) to detect abnormal behaviors in the network. Many available datasets were created to evaluate the performance of the model, such as KDD CUP99 and NSL-KDD. However, with the increasing scale of data and the emergence of advanced attacks, conventional machine learning algorithms can hardly perform well. Fortunately, the development of deep learning provides new direction for solving these problems. In this paper, in order to detect novel attacks in a network and improve detection efficiency, we proposed a flexible framework based on deep neural network (DNN). In our framework, we apply different feature reduction methods and activation functions to get the best performance. Moreover, through changing hyper-parameter of the model, we select better network structure. To evaluate our framework, we select ISCX 2012 and CICIDS 2017 as a benchmark and apply the proposed framework to these datasets. As a result, we observe high accuracy rate and low FAR for both binary and multi-class classifications. Overall, our proposed framework is universal and useful for detecting zero-day attacks.

Liu Zhiqiang, Lin Zhijun, Gong Ting, Shi Yucheng, Mohi-Ud-Din Ghulam
Program Execution Comprehension Modelling for Algorithmic Languages Learning Using Ontology-Based Techniques

In this paper, we propose an ontology-based approach to model a program execution comprehension so to be able to explain to the novice programmer the essence of his/her error. We have studied the algorithmic languages model operating with actions and basic control structures (“sequence,” “branching,” and “looping”) and designed the rules to capture any deviation from the permissible.

Mikhail Denisov, Anton Anikin, Oleg Sychev, Alexander Katyshev
Multipathing Communication in Heterogeneous Ad Hoc Network

Multipath routing is a solution that meets the requirements of existing systems (security, bandwidth, network lifetime, etc.). Currently, several studies are trying to apply and adapt the principle of multipath in several types of networks for benefiting from its advantages. The wireless network without infrastructure (ad hoc network) is more flexible and less expensive compared to other networks and uses in its turn the multipath routing in the communication between the devices to enhance its strengths. The ad hoc network regroups several categories of network, and each one has its own node characteristics and communication protocol. In most application areas based on new technologies, such as smart cities, precision agriculture, medicine, and others, we can find different types of ad hoc networks in the same architecture constituting a heterogeneous network. So, the heterogeneous ad hoc network (HAdN) appears and brings with it a new line of research concerning the common and compatible routing protocols inside this type of network. This article proposes a new multipath routing protocol version based on the OLSR standard ensuring the direct communication between the various HAdN components. The new protocol will initially be implemented in a specific network architecture consisting of MANET, VANET, and FANET.

Chaimae Benjbara, Nada Mouchfiq, Ahmed Habbani
Integration of Web-Scraped Data in CPM Tools: The Case of Project Sibilla

Modern corporate performance management (CPM) systems are crucial tools for enterprises, but they typically lack a seamless integration with solutions in the Industry 4.0 domain for the exploitation of large amounts of data originated outside the enterprise boundaries. In this paper, we propose a solution to this problem, according to lessons learned in the development of project “Sibilla,” aimed at devising innovative tools in the business intelligence area. A proper software module is introduced with the purpose of enriching existing predictive analysis models with knowledge extracted from the Web and social networks. In particular, we describe how to support two functionalities: identification of planned real-world events and monitoring of public opinion on topics of interest to the company. The effectiveness of the proposed solution has been evaluated by means of a long-term experimental campaign.

Alessio Bechini, Beatrice Lazzerini, Francesco Marcelloni, Alessandro Renda
Security Issues in MANETs: A Survey

Recently, the notion of security has become increasingly important to research on wireless networks, later the Internet of things (IoT) and ad hoc networks that are a part of IoT environment. The various changes in the digital environment, as it must identify the different changes in the digital environment, are namely the protection of the personal and professional data of this intelligent world. In the work already done, there are several methods and levels of security. In this study, we identify that security aggression MANETs are facing with the security services to provide in each layer. In order to reach our target, we did a literature search gathering information on different types of attacks and solutions, as well as the latest security work. Then, we discuss the proposed solutions to ensure a high level of security in networks by recalling the latest work of our team. Finally, we propose a method of security better adapted to our needs based on blockchain technology.

Nada Mouchfiq, Ahmed Habbani, Chaimae Benjbara
Depression Detection Using Audio-Visual Data and Artificial Intelligence: A Systematic Mapping Study

Major depression disorder is a mental issue that has been increasing in the last decade, in consequence, prediction or detection of this mental disorder in early stages is necessary. Artificial intelligence techniques have been developed in order to ease the diagnosis of different illnesses, including depression, using audio-visual information such as voice or video recordings and medical images. This research field is growing, and some organizations and descriptions are required. In the present work, a systematic mapping study was conducted in order to summarize the factors involved in depression detection such as artificial intelligence techniques, source of information, and depression scales.

José Balbuena, Hilda Samamé, Silvana Almeyda, Juan Mendoza, José Antonio Pow-Sang
IT&C System Solution for Visually Impaired Romanian Teenagers

The system presented in this paper is mainly aimed for improving the quality of education. It is intended to help people with visual, reading–writing disabilities, learning disabilities, and moderate autism spectrum disorders. The system would assist them in social integration as it is well known that such people are isolated and hence other health problems arise over time. In Romania, there is high speed and fast Internet connection, enabling user’s access from almost any place of the country. This represents the background for access to education that could start with discovering the keyboard with voice assistance, browse the Internet with voice assistance, also they could learn to use different voice-assisted programs in their native language or choose any other language, thus opening new horizons for education, knowledge, and information. The solution is based on the Vinux-GNU/Linux operating system, which has undergone multiple adaptations, designed to facilitate user access to information even when it has an older generation computer, so no new financial investments are needed.

Isabela Todirițe, Corina Radu Frenţ, Mihaiela Iliescu
A Survey on Named Entity Recognition Solutions Applied for Cybersecurity-Related Text Processing

Named entity recognition (NER) is one of the most common Natural Language Processing (NLP) tasks. As nowadays large quantities of unstructured data are produced, the organizations have begun to be more interested in NER solutions. In the first part of this article, we describe the evolution of NER, and we discuss the most common NER approaches. Later, we address the state-of-the-art NER machine learning solutions. We focus both on open-source and commercial solutions. The most important solutions are identified and compared based on a methodology proposed by the authors. Since the authors are involved in using NER on cybersecurity-related text, the study focuses mainly on NER aspects related to cybersecurity domain. Nevertheless, this survey has a general nature, and therefore, our conclusions can be useful as well for those interested in using NER solutions in other domains.

Tiberiu-Marian Georgescu, Bogdan Iancu, Alin Zamfiroiu, Mihai Doinea, Catalin Emilian Boja, Cosmin Cartas
URL’s Folder Name Length as a Phishing Detection Feature

Phishing is a cybercrime in which, the phishers try to control users’ credentials. The phishers usually construct fake URLs that take to phishing websites where the users might disclose, and loss their credentials. This paper introduces a new feature to detect phishing URLs. The length of folder name of URL’s path is utilized in this work as a phishing detection feature, namely FldrNmLnth. Results from analysis of two email datasets show that many of phishing URLs are constructed using upto 230 characters as names of their folders’ names. The length of folders’ names of legitimate URLs, on the other side, has not exceeded 30 characters. This length variance motivates the use of folder name length as a feature to detect phishing URLs.

Melad Mohamed Al-Daeef, Nurlida Basir, Madihah Mohd Saudi
Future of Governance for Digital Platform Ecosystems

The future of governance and regulation of digital platforms is uncertain. The development of digital platforms depends on a number of political and social factors in addition to economic and technological drivers. Particularly, the role of governance and institutions are crucial for structuring platform ecosystems. Therefore, this paper explores alternative futures of digital platform ecosystem governance through the scenario planning approach. This approach allows considering alternative future trajectories rather than rely on extrapolation of current trends. The paper discusses different scenarios developed by international and national organizations. Three alternative futures for the governance of digital platform ecosystems emerge resulting from scenario matching: private platform ecosystems, government platform ecosystems, and decentralized platform ecosystems. The conclusion highlights the key implications of different meta-scenarios.

Meelis Kitsing, Johanna Vallistu
Azimuth Tree-Based Self-organizing Protocol for Internet of Things

An azimuth tree-based network algorithm is proposed on a uniform random distribution of scattered points which is actually graph-based. In the proposed azimuth on a tree-based network, first tree is established through the algorithm for finding the weight function of the tree would constrain the number of points to be considered during azimuth routing, thereby limiting the search space, reducing time complexities, and inducing further optimizations. The idea behind our paper would be applying the azimuth algorithm on a selected number of points, which would be achieved through the efficient tree-based routing protocol. So, firstly, we would apply our tree-based protocol on the uniform random distribution of points. The limited number of points output by this algorithm would be fed as input to azimuth routing. And in the protocol proposed by us, the aggregation of data in this tree-based network helps in reducing the network load and the energy consumption. And this research work mainly revolves around the criteria to be taken into consideration for balancing factors like these and construct a tree-based network which is better in terms of both lifetime of the network and the successful routing of protocols. With the help of simulation implemented using C++ and then constructing its corresponding graph, we have shown how considering all the factors possible has improved the performance of tree-based network. And finally, a comparative analysis is done where our proposed model is compared to the already existing traditional routing protocols namely AODV, DSDV, LEACH, Azimuth-based algorithm.

Veena Anand, Prerana Agrawal, Pothuri Surendra Varma, Sudhakar Pandey, Siddhant Kumar
Evaluation of Confidentiality and Data Integrity Based on User Session for IoT Environment

Confidentiality and data integrity are two important security services that are provided by the most popular VPN protocols such IPsec and SSL/TLS. It is a well-known fact that security comes at the cost of performance, and performance is affected by the cryptographic algorithms and their execution. In this study, confidentiality and data integrity processes were implemented using client--server Java applications in a real LAN to assess their performance based on different encryptions methods. The execution time of data encryption, data decryption, and data integrity verification was measured and total session times were computed for the following encryption algorithms AES, Blowfish, 3DES, RC2, MD5, and SHA-1. The observation results have been analyzed and the performance of these encryption algorithms and digest ciphers was compared, interpreted, and outlined in terms of total session time and different security parameters (data integrity, encryption, decryption, and data integrity verification).

Alwi Bamhdi, Omar Abdulgadir, Akram Kargar, Ahmed Patel, Bandar Al Ghamdi
Quantales for Uncertainty in Design Engineering

In this paper we provide recommendations on how to use quantales as algebraic structures to represent uncertainty and many-valuedness in design engineering using relational views for connecting and combining information. Information is further detailed as based on underlying signatures of types and operators, providing expressions and terms that also become subjected to many-valued qualifications. Machine and engineering design, and related design structures usually adopt rather trivial relational models, and shallow expressions to describe various conditions. In particular, uncertainty, e.g., in prediction and risk estimation, is often based on quite rudimentary and ad-hoc probabilities of events that are mostly just named rather than described in detail. The objects in question being just named items, without elaborating on the internal structure of these objects, makes these descriptions to be simple constants, and truth valuation remains as only binary. We will show how objects can be structured, and how structured objects can be related using various algebraic structures. This enables to provide a richer model also on many-valuedness from a logical point of view. Specifically we will look at the algebraization of the Design Structure Matrix (DSM).

Patrik Eklund
Deciphering of the gpsOne File Format for Assisted GPS Service

The paper is concerned with deciphering the data format of a gpsOneXTRA binary file for A-GPS web service. We consider mandatory data content of the file and reveal the changes of this content at different moments of time. The frequency of the changes hints on the location of records for current GPS date and satellite orbits information. Comparing the repeating data patterns against reference orbits information, we obtain meaning of data fields of the orbit record for each operational satellite. The deciphered file header and GPS almanac data layout are provided as tables within the paper.

Vladimir Vinnikov, Ekaterina Pshehotskaya
Learning Effects of Different Learning Materials About Automated Driving Level 3: Evidence from a Propensity Score Matching Estimator

There are always big issues to use new technology. The users need to know how to use it and what will happen if they do not use it properly because of lack of knowledge. Automated vehicles equivalent to driving automation levels 3 and 4 and advanced driving support systems in the levels are the same. Necessary knowledge and information to be acquired by drivers and pedestrians as well as effective educational methods should be identified. Furthermore, there are individual differences what one needs to know about automated driving as a driver and how to relate to an automated vehicle as a driver. In this research, safe driving education prototype contents were developed that can absorb differences in personal attributes such as learning styles, ages, and personal traits. This study examined how the traffic safety education, such as safety training for driving schools, should be provided as effective educational methods.

Maki Arame, Junko Handa, Yoshiko Goda, Masashi Toda, Ryuichi Matsuba, Huiping Zhou, Makoto Itoh, Satoshi Kitazaki
A Health Decision Support Framework for the Prediction of Cognitive Ability in Older People

Deterioration in cognitive functioning has become a serious public health burden due to aging of the population. Strategies for maintaining cognitive abilities with age are needed critically. The objectives of this study were: first to build a system that could accurately predict reduced cognitive function in older adults and; second, through this, to identify features that predict reduced cognitive function. Three tests of cognitive ability were investigated using data from the English Longitudinal Study of Aging (ELSA). Six machine learning algorithms were separately implemented in the system and their performance was compared in terms of the three cognitive tests. For each cognitive test, potentially important risk factors were identified as protective factors against cognitive aging. The findings from this study enhance our understanding of the underlying mechanisms that affect cognitive aging.

Hui Yang, Peter A. Bath
Incorporating Stability Estimation into Quality of Service Routing in an SDN-Based Data Centre Network

Software defined networking (SDN) has emerged as effective paradigm in which the working principle was based on the separation of data plane from the control plane. This approach has proved to be quite advantageous in terms of its flexibility and programmability, thus enhancing better service provisioning and innovative network management. Despite the high results and improved expectations from the approach, one of the major challenges that the system faces is the level of network stability that the approach offers. We then tend to opined that based on our findings in literature, only few articles have addressed the level of network stability that SDN paradigm is offering as well as its improvements. Several approaches include basically the deployment of efficient routing protocols to ensure that the network attains some level of resilience. Even though the routing plays important role in the network environment, the control of the flow setup connection between the switches and controllers helps to instill better packet delivery with avoidance of network failure. We introduced a dynamic mapping orchestrator engine into the network setup which helps to unburden the controllers, as well determine the need for switching to either single, double or multiple mapping of switches to controllers. We evaluated the proposed approach by determining the average flow setup time, the QoS produced in terms of throughput and the fairness of the switching between the mappings of switches to controllers. The stability of the proposed approach was estimated to have been improved by 22% over the existing approach.

Ayotuyi T. Akinola, Matthew O. Adigun, Pragasen Mudali
Convergence Speed up Using Convolutional Neural Network Combining with Long Short-Term Memory for American Sign Language Alphabet Recognition

In sign language alphabet recognition problem, the scope of study limits only static hand gestures which not cover all gestures of sign language. This paper aims to find an approach for recognizing the static and dynamic gestures of American Sign Language (ASL) alphabet and apply GANs to generates synthetic images to increase dataset size. The proposed method combines convolutional neural networks (CNN) with long short-term memory (LSTM) networks to extract the features and classify images of the American Sign Language alphabet along various dimensions. With two consecutive images, this proposed method has an accuracy of over 97% and on 1D vector images, accuracy reaches 90% in large batch size when were tested on various batch sizes and epochs. Thus, this method is more appropriate for two consecutive images than on 1D vector images. For dynamic features, the performance of the proposed CNN-LSTM on two consecutive images is lower than the simple CNN at the beginning epoch, but the accuracy converged quickly, and finally, it reaches to the accuracy of simple CNN in a few epochs. Our proposed approach offers good results and better than simple CNN for dynamic ASL alphabet gestures, especially on 1D vector images.

Busarakorn Supitchaya, Varin Chouvatut
Pain Detection Using Deep Learning with Evaluation System

Recent evidence has appeared that major enhancements in patient results can be increased by the periodical observing patient pain levels by medical staff in hospitals. Nevertheless, owing to the responsibility and pressure that the staffs have, this kind of observation has been complicated to withstand; thus, a system that works automatically could be the solution. Using an automatic facial expression system to detect pain which pain can be defined via several facial action units (AUs). To simplify pain detection using deep learning, data were collected from the UNBC database, which contains sequences of images that show participants’ faces while they were doing an arrangement of movement-of-motion tests. To improve pain detection using facial expressions, this research proposes a pain detection technique that uses deep learning. Finally, the resulting of the experiment will be compared with the self-reports, and doctors will be asked to evaluate the system.

Kornprom Pikulkaew, Ekkarat Boonchieng, Waraporn Boonchieng, Varin Chouvatut
Cooperative Location Acquisition of Mobile Nodes Without Direct Distance Measurement

In wireless networks composed of numbers of mobile wireless nodes, their location information is required to be achieved for fundamental network services such as routing of data messages and determination of server nodes providing various services such as name services and for supporting network applications based on locations of the mobile wireless nodes, for instance, sensor network applications and ITS applications. Until now, various methods for achieving distances between two mobile wireless nodes have been proposed; however, some of them requires too expensive devices and others achieves too low-resolution results. Most of the methods using RSSI of transmitted wireless signals between two mobile wireless nodes cannot provide enough high resolution due to large deviation of transmission delay and the multi-pass problem. This paper proposes a novel method for achieving distances between mobile wireless nodes by cooperation of 3 mobile wireless nodes with cameras and a device for measuring its own migration length. The cameras are only used for measuring angles with high accuracy between 2 mobile wireless nodes which are independent of their shapes, sizes and distances.

Masaaki Namekata, Hiroaki Higaki
E-Learning and Industry 4.0: A Chatbot for Training Employees

Within what is called the Fourth Industrial Revolution, one of the problems that companies encounter frequently, in order to ensure themselves and their products and services over time, is the need for continuous training of the people who work in synergy for them. If continuing education is a problem, e-learning is its natural solution. The research concerns the realization of a system capable of providing a constant, reliable, and friendly help through a practical and nice chatbot based on context processing. In particular, the proposed chatbot acts as a reminder and follows the user during his personal corporate training, ready to provide, when needed, the necessary and useful teaching material to complete the educational course.

Fabio Clarizia, Massimo De Santo, Marco Lombardi, Domenico Santaniello
A Secure Smart City Infrastructure Framework for E-Service Delivery Within a Developing Country: A Case of Windhoek in Namibia

As cities embrace technologies, they become more intelligent ubiquitous and unlimited connection. This is enabled by fast broadband and better supporting infrastructure. African cities are making effort to improve service delivery using technologies. As these developments are witnessed, information sharing, infrastructure requirements and security issues become very common. There are a number or study papers on smart cities; however, less has been studied on the smart city security and Infrastructure. Since it is clear that within a smart city, high Internet speed, intelligent systems efficient services are crucial, there is need for a secure supporting information infrastructure. Qualitative data was gathered from participants within a case study approach. Results should the current ICT initiatives to support smart cities. At the same time, all participants engaged agree that a secure infrastructure framework is key for African cities to become smart. The paper presents a comprehensive Secure Smart City Infrastructure Framework (SSCIF) for the City of Windhoek in Namibia. The framework aims to enable e-service delivery within a smart city.

Licky Richard Erastus, Nobert Rangarirai Jere, Fungai Bhunu Shava
A Multigraph Approach for Supporting Computer Network Monitoring Systems

The pervasiveness of information technologies has reached very high levels: most human activities involve the use of sensor-based systems connected to the network. The increasingly widespread use of the Internet of things has significantly improved our quality of life but has introduced a series of new problems, especially from the security point of view. Protecting these systems from cyber-attacks has become a priority as possible malfunctions can lead to issues with a significant social impact. Imagine, for example, computer attacks on smart cars connected to the network or remotely controlled electrical or water systems. Protecting this type of system is a complex task as there are many elements to consider and the data to be monitored. An analysis able to foresee eventual attacks through the study of the data and their variations could be a useful tool to prevent malfunctions. This paper proposes a methodology based on the integrated use of three graphic models to address the problem of preventing attacks on pervasive systems from three different perspectives: probabilistic, contextual, and ontological. The paper proposes the use of Bayesian networks built through an ontological definition of the problem dropped on a particular context represented by a Context Dimension Tree—the proposed approach experiments in a real scenario providing satisfactory results.

Francesco Colace, Muhammad Khan, Marco Lombardi, Domenico Santaniello
Comparing Phonetic Reduction in Russian as First and Second Language: From Psycholinguistics to Natural Language Processing

In the paper, we argue that it is necessary to collect and analyze casual speech of the speakers from different age groups, both native speakers and those who study a language as a second one, in order to understand the mechanisms of spoken word production and recognition and to improve current automatic systems of natural language processing. We provide a short overview of the corpora of adult, children and adolescent Russian speech we develop and then focus on the methodology and results of a study of phonetic reduction in the speech of 16 Chinese students learning Russian as a foreign language. We found out similar tendencies of phonetic reduction in the speech of the Chinese students and in the speech of native speakers of Russian. At the same time, the speech of Chinese students, unlike the speech of Russian-speaking children aged four to six years, is characterized by a large number of examples with sound changes. The second language learners of Russian usually have different realizations of one and the same word in their speech. The results we obtained can be used for spoken word recognition modeling, as well as for various educational purposes including teaching Russian as a foreign language.

Elena I. Riekhakaynen
Ensuring Continued Use of Information Systems in an Organization

The information systems (IS) play a significant role in an organization’s achievement, especially in digital era. IS can accelerate organization thru information handling and give substantial affects of IS success and later to its continuous use. The growth in today’s business in terms of technology and commercial enterprise models requires comprehensive elements to support usage and continuance use of IS. Therefore, this research has a look to develop an IS continuance version through the continuation of expectation confirmation mode (ECM) by means of integrating new factors from other related theories. The proposed model is evaluated using questionnaire among contributors of the user of endowment firms that use e-endowment system. The feedback from the reliable respondents had been analysed using the partial least squares (PLS) primarily based on structural equation modelling (SEM) technique. The results show that support, usage, and technology suitability complement behavioural intention elements to IS continuance intention. The results of this study contributed to the new understanding in IS continuance environment and gives a possibility for developing an influential plan of IS continuance inside the organizations.

Mohd Zuhan Mohd Zain, Ab Razak Che Hussin
A Cost-Effective 3D Acquisition and Visualization Framework for Cultural Heritage

Museums and cultural institutions, in general, are in a constant challenge of adding more value to their collections. The attractiveness of assets is practically tightly related to their value obeying the offer and demand law. New digital visualization technologies are found to give more excitements, especially to the younger generation as it is proven by multiple studies. Nowadays, museums around the world are currently trying to promote their collections through new multimedia and digital technologies such as 3D modeling, virtual reality (VR), augmented reality (AR), and serious games. However, the difficulty and the resources required to implement such technologies present a real challenge. Through this paper, we propose a 3D acquisition and visualization framework aiming mostly at increasing the value of cultural collections. This framework preserves cost-effectiveness and time constraints while still introducing new ways of visualization and interaction with high-quality 3D models of cultural objects.

Hosameldin Osman Ahmed, Abdelhak Belhi, Taha Alfaqheri, Abdelaziz Bouras, Abdul H. Sadka, Sebti Foufou
Practical Vulnerability Analysis of Mouse Data According to Offensive Security Based on Machine Learning

We demonstrate a security threat of mouse data by differentiating the real mouse data from the dummy mouse data by deriving features to have high accuracy based on data science. Features appearing between the mouse coordinates input by the user are analyzed, and the feature is defined as a feature for machine learning models to derive a method of improving the accuracy. As a result, we found a feature where the distance between coordinates is concentrated in a specific range. When the distance is used as a feature, we verified that the mouse data is stolen more accurately.

Kyungroul Lee, Hoon Ko, Hyoungju Kim, Sun-Young Lee, Junho Choi
Preliminary Study and Implementation of Chiang Mai Tourism Platform Based on DOSA

To cope with data complication systematically, a tourism platform is proposed through a newly-defined concept named Data Oriented Security Architecture or DOSA. The framework of DOSA shifts an attention from “Application” in a conventional method to “Data”. The main idea is to neutralize the numerously different sources of data into single entity called Data Register Center and strengthen the security of data by utilizing the public- and private-key encryption. Thus it could be simply adaptable to the varieties of applications and easily extensible to sustainable development. The advantages of DOSA include (1) maintainability of data ownership, (2) improvement in security and privacy, and (3) reduction in storage that, as a result, are suitable to modern application development. The application prototype is implemented in web application and piloted by a set of tourist destination in Chiang Mai. This work contains the implementation from Data Register Center in DOSA through the Tourism Service Platform. Although the sample does not include all features of the DOSA, the results are comprehensive and ready for the future extension and improvement.

Vorada Panpeng, Miao Fang, Aniwat Phaphuangwittayakul, Tanarat Rattanadamrongaksorn
Factors of Implementing Citizen-Centric e-Government in Developing Countries: Namibia

The low adoption and use of e-Government have motivated interest in researching citizen-centric e-Government. This is important to developing countries where e-Government is seen as promoting social exclusion due to differences in income, access to ICTs and literacy rates across the populace. This study used a prototype m-Government application to investigate factors of citizen-centric e-Government. The study targeted one of the Ministries in Namibia and investigated challenges faced by citizens when accessing government service. While m-Government and e-Government could play a key role in addressing some of the challenges faced, it was noted that the government need to strategically implement e-Government given a number of contextual factors. For instance, while an urban-based citizen could afford a smartphone and operate one, the case is different with a citizen living in rural communities; rural populace, characterised by the aged cannot afford smartphones nor operate ones without assistance. This is mainly down to the fact that part of the rural populace cannot read and write in English or their native language. Hence, this study suggests understanding ICTs accessible to the populace, ICT skills, attitude towards technology, ICT infrastructure, costs, security and using a community-based partnership approach as key factors of e-Government in Namibia.

Karin Frohlich, Marko Nieminen, Antti Pinomaa
Integration of Blockchain and Remote Database Access Protocol-Based Database

Many companies are relying on software to manage their businesses. Usually, the software, especially those used by smaller companies, is not secure against unauthorized or unethical data manipulation on the database level. This paper recommends and demonstrates the use of blockchain for securing small enterprises against hacking by alerting the management whenever a change is made to the data without using the authorized channels. This is done through blockchain technology’s inherent hash replication and mining algorithm. The paper shows an application where this idea has successfully been implemented for a desktop application built upon a remote database access (RDA) protocol-based relational database management system (RDBMS) such as remote MySQL and remote Oracle database.

Muhammad Jafar Sadeq, S. Rayhan Kabir, Marjan Akter, Rokeya Forhat, Rafita Haque, Md. Akhtaruzzaman
Crossing the Artificial Intelligence (AI) Chasm, Albeit Using Constrained IoT Edges and Tiny ML, for Creating a Sustainable Food Future

Big data surrounds us. Every minute, our smartphone collects huge amount of data from geolocations to next clickable item on the e-commerce site. Data has become one of the most important commodities for the individuals and companies. Nevertheless, this data revolution has not touched every economic sector, especially rural economies; e.g., small farmers have largely passed over the data revolution, in the developing countries due to infrastructure and constrained compute environments. Not only this is a huge missed opportunity for the big data companies, but it is one of the significant obstacles in the path toward sustainable food and a huge inhibitor closing economic disparities. The purpose of the paper is to develop a framework to deploy artificial intelligence models in constrained compute environments that enable remote rural areas and small farmers to join the data revolution and start contribution to the digital economy and empower the world through the data to create a sustainable food for our collective future. In a nutshell, close the digital gap by crossing the AI chasm to democratize AI for poor and helpless farmers and help ourselves by creating sustainable food future.

Chandrasekar Vuppalapati, Anitha Ilapakurti, Sharat Kedari, Raja Vuppalapati, Jaya Vuppalapati, Santosh Kedari
Design and Development of System for Post-infection Attack Behavioral Analysis

Attacks targeting government and public networks for getting political, and strategic advantage is a common phenomenon. Most of the time these targeted attacks are state-sponsored and banks heavily on the social engineering techniques to get a successful foothold in the targeted network. As these attacks use a combination of social engineering tactics along with zero-day exploits, their detection using conventional signature-based detection systems is a challenge. The first step towards the mitigation of these targeted attacks is to develop a mechanism to analyze and characterize these attacks. This study proposed a conceptual framework for capturing, collection, and analyses of targeted attacks in an enterprise network scenario. This framework provides a mechanism for capturing and performing post-infection behavioral analysis on the captured adversary.

Toivo Herman Kamati, Dharm Singh Jat, Saurabh Chamotra
Multi-threaded Multilayer Neural Network for Character Recognition

New techniques are evolving day by day to recognize characters in any image, text, cards, etc. Optical character recognition is a well-accepted technique to recognize characters in scene or optically captured image. In this present work, we have evolved new method which employs feed forward back propagation (FFPB) technique interlaced with multi-threaded multilayer perceptron neural network (MTMLNN) to recognize the characters with speedup in operations and better accuracy in character recognition. In present methodology, we have employed structured similarity index measure (SSIM) in MTMLNN to compare image and sigmoid function for neuron activation with stochastic gradient descent method for optimization, to train the neural network to recognize the input character with better accuracy.

Arun Vaishnav, Manju Mandot, Priyanka Soni
Threat Detection in Social Media Images Using the Inception-v3 Model

Threat detection in images within social media content has become an important aspect of content monitoring. This task can be achieved using various image object detection and classification techniques. Recently, object detection has become an important task of machine learning, with significant studies dedicated to using deep learning techniques, especially the convolutional neural network (CNN), in the computer vision field. The current study involves an experiment using transfer learning technology to retrain the Inception-v3 model from TensorFlow in terms of the collected dataset related to known threat content. The novelty of this research work lies in the threat detection of images shared on social media which was not addressed before. The model achieves a high accuracy of around 96% in threat detection. The results of this research will be helpful in monitoring and tracking social media image content in terms of the detection of threats in the images shared among users, while the system can be used as a standalone system or as part of larger systems.

Shatha AbdulAziz AlAjlan, Abdul Khader Jilani Saudagar
Backmatter
Metadata
Title
Proceedings of Fifth International Congress on Information and Communication Technology
Editors
Prof. Dr. Xin-She Yang
Prof. Simon Sherratt
Dr. Nilanjan Dey
Amit Joshi
Copyright Year
2021
Publisher
Springer Singapore
Electronic ISBN
978-981-15-5859-7
Print ISBN
978-981-15-5858-0
DOI
https://doi.org/10.1007/978-981-15-5859-7