Natural Language Processing for Software Engineering (eBook)

Rajesh Kumar Chakrawarti, Ranjana Sikarwar, Sanjaya Kumar Sarangi, Samson Arun Raj Albert Raj, Shweta Gupta, K. Sakthidasan Sankaran, Romil Rawat (Herausgeber)

eBook Download: EPUB

2025
803 Seiten
Wiley-Scrivener (Verlag)
978-1-394-27244-0 (ISBN)

Infos

Discover how Natural Language Processing for Software Engineering can transform your understanding of agile development, equipping you with essential tools and insights to enhance software quality and responsiveness in today's rapidly changing technological landscape.

Agile development enhances business responsiveness through continuous software delivery, emphasizing iterative methodologies that produce incremental, usable software. Working software is the main measure of progress, and ongoing customer collaboration is essential. Approaches like Scrum, eXtreme Programming (XP), and Crystal share these principles but differ in focus: Scrum reduces documentation, XP improves software quality and adaptability to changing requirements, and Crystal emphasizes people and interactions while retaining key artifacts. Modifying software systems designed with Object-Oriented Analysis and Design can be costly and time-consuming in rapidly changing environments requiring frequent updates. This book explores how natural language processing can enhance agile methodologies, particularly in requirements engineering. It introduces tools that help developers create, organize, and update documentation throughout the agile project process.

Rajesh Kumar Chakrawarti, PhD, is a dean and professor in the Department of Computer Science and Engineering at Sushila Devi Bansal College, Bansal Group of Institutions, India. He has over 20 years of professional experience in academia and industry. Additionally, he has organized and attended over 200 seminars, workshops, and conferences and has published over 100 research papers and book chapters in nationally and internationally revered publications.

Ranjana Sikarwar is currently pursuing a PhD from Amity University, Gwalior. She completed her Bachelor of Engineering in 2006 and Master of Technology in Computer Science and Engineering in 2015. Her research interests include social network analysis, graph mining, machine learning, Internet of Things, and deep learning.

Sanjaya Kumar Sarangi, PhD, is an adjunct professor and coordinator at Utkal University with over 23 years of experience in the academic, research, and industry sectors. He has a number of publications in journals and conferences, has authored many textbooks and book chapters, and has more than 30 national and international patents. He is an active member and life member of many associations, as well as an editor, technical program committee member, and reviewer in reputed journals and conferences. He has dedicated his career to advancing information and communication technology to enhance and optimize worldwide research and information dissemination, leading to improved student learning and teaching methods.

Samson Arun Raj Albert Raj, PhD, is an assistant professor and placement coordinator in the Division of Computer Science and Engineering, School of Computer Science and Technology, Karunya Institute of Technology and Sciences, Tamil Nadu, India. His research is focused on smart city development using drone networks and energy grids with various applications, and his areas of expertise include wireless sensor networks, vehicular ad-hoc networks, and intelligent transportation systems.

Shweta Gupta is an assistant professor in the Computer Science and Engineering Department at Medicaps University, Indore (M.P.), India. She focuses on natural language processing, data mining, and machine learning. She aims to close the knowledge gap between theory and real-world applications in the tech sector through her passion for research and teaching. Her approach centers on encouraging creativity and motivating students to strive for technological excellence.

Krishnan Sakthidasan Sankaran, PhD, is a professor in the Department of Electronics and Communication Engineering at Hindustan Institute of Technology and Science, India. He has been a senior member of the Institute of Electrical and Electronics Engineers for the past ten years and has published more than 70 papers in refereed journals and international conferences. He has also published three books to his credit. His research interests include image processing, wireless networks, cloud computing, and antenna design.

Romil Rawat has attended several research programs and received research grants from the United States, Germany, Italy, and the United Kingdom. He has chaired international conferences and hosted several research events, in addition to publishing several research patents. His research interests include cybersecurity, Internet of Things, dark web crime analysis and investigation techniques, and working towards tracing illicit anonymous contents of cyber terrorism and criminal activities.

Textauszug

1
Machine Learning and Artificial Intelligence for Detecting Cyber Security Threats in IoT Environmment

Ravindra Bhardwaj1*, Sreenivasulu Gogula2, Bidisha Bhabani3, K. Kanagalakshmi4, Aparajita Mukherjee5 and D. Vetrithangam6

1Deparment of Physics and Computer Science, Dayalbagh Educational Institute (Deemed to be University), Agra, Uttar Pradesh, India

2Department of CSE (Data Science), Vardhaman College of Engineering, Shamshabad, Hyderabad, India

3Department of Computer Science and Engineering, University of Engineering and Management (UEM), New Town, West Bengal, India

4Department of Computer Applications, SRM Institute of Science and Technology (Deemed to be University), Trichy, India

5Department of Computer Science and Engineering, Institute of Engineering and Management, University of Engineering and Management (UEM), New Town, Kolkata, West Bengal, India

6Department of Computer Science & Engineering University, Institute of Engineering, Chandigarh University, Mohali, Punjab, India

Abstract

The Internet of Things (IoT) refers to the increasing connectivity of many human-made entities, such as healthcare systems, smart homes, and smart grids, through the internet. Currently, a vast amount of material and expertise has been widely spread. These networks give rise to several security threats and privacy concerns. Intrusions refer to malevolent and unlawful actions that cause harm to the network. IoT networks are susceptible to a diverse range of security issues due to their widespread presence. Cyber attacks on the IoT architecture can lead to the loss of information or data, as well as the sluggishness of IoT devices. For the past twenty years, an Intrusion Detection System has been utilized to ensure the security of data and networks. Conventional intrusion detection technologies are ineffective in detecting security breaches in the Internet of Things (IoT) because of the distinct standards and protocol stacks used in its network. Regularly analyzing the vast amount of data created by IoT is a tough task due to its endless nature. An intrusion detection system (IDS) is employed to safeguard a system or network against unauthorized access by actively monitoring and identifying any potentially malicious or suspicious activities. Machine learning technologies provide robust and efficient approaches for mitigating these distinct hazards. The establishment of a robust machine learning system is the key to acquiring networks that are free from any form of threats.

Keywords: Machine learning, Internet of Things, security, privacy, attacks, vulnerability, intrusions

1.1 Introduction

The use of connected devices made ordinary chores easier and more efficient. They also provide a lot of information that is of great use. Connected automobiles, for example, may be able to take use of services that provide driver assistance. Medical devices give detailed patient records. The unfortunate reality is that a digital assault is possible on any device that is capable of establishing a connection to the internet. In worst case, many of these devices are missing even the most basic safety safeguards. According to the authors of the report, almost all of the data flow associated with the internet of things (98%) is not secured. This information may be obtained by anybody with little effort. To repeat, devices that are connected to the Internet of Things provide fraudsters with an easy target. Not only might their information be stolen, but perhaps other sensitive data as well. Using one of these devices is a frequent strategy used by hackers to gain access to a company’s internal network. The sheer number of these devices and the settings they control may be enough to pique the interest of a cyber-attacker [1] as given in Figure 1.1: Increasing Number of DDOS Attacks [Source: Cisco Annual Internet Report 2018-2023] and in Figure 1.2: Threats to Internet of Things.

In a smart environment, any number of items, including databases of user credentials, electronic sensors, CCTV installations, access controls, personal electronic devices, recorded biometrics, and so on, might be the target of an attack. It is essential to protect the confidentiality, integrity, availability, authentication, and authorization features of the IoT architecture from a security point of view [2]. DDoS attacks are becoming more common, and Cisco’s Annual Internet Report (2018-2023) White Paper forecasts that the total number of DDoS attacks would more than double from the 7.9 million that were seen in 2018 to anywhere over 15 million by 2023 as shown in Figure 1.1.

Figure 1.1 Increasing number of DDOS attacks [Source: Cisco Annual Internet Report 2018-2023].

Figure 1.2 Threats to Internet of Things.

According to the survey, 57% of IoT devices that are connected via this insecure traffic are susceptible to medium- to high-severity attacks, making them an easy target for cybercriminals [3]. In addition, the survey found that 41% of attacks target IoT vulnerabilities by scanning them against publicly available databases of known security flaws. The analysis is shown in Figure 1.2.

According to the Internet of Things Threat Report published by Palo Alto Networks in March 2020, 98% of all traffic from IoT devices is unencrypted, giving attackers a chance to eavesdrop. This network contains sensitive and private information that is easily accessible to attackers, who may then sell the information on the dark web for a profit.

1.2 Need of Vulnerability Identification

Vulnerabilities in IoT network are increasing every year. As shown in Figure 1.3, IoT environment is experiencing, a large number of new vulnerabilities every year. All the Internet of Things applications—smart city, smart farming, smart healthcare, smart transportation, and smart traffic—are experiencing new vulnerabilities and increasing number of attacks every year. Also, vulnerabilities and attacks are increasing every year. Number of vulnerabilities has increased threefold in the last decade and twofold in last five years as represented in Figure 1.3: Number of New Vulnerabilities Identified in IOT [Source- IBM X-Force Threat Intelligence Index 2022].

Figure 1.3 Number of new vulnerabilities identified in IoT [Source- IBM X-Force Threat Intelligence Index 2022].

The process of determining how vulnerable a system is to attack is referred to as a vulnerability scan. This kind of scan is carried out to identify potential entry points into a computer or network so that appropriate preventative measures may be taken. Automated scanning methods check applications to see if they have any security problems to establish whether or not there are vulnerabilities in an organization’s internal network. Users are spared the time and effort required to carry out hundreds or even thousands of manual tests for each kind of vulnerability since vulnerability scanners automate the process of searching for security issues in a system.

To maintain the integrity of the system’s protections, it is essential to assign vulnerabilities a severity ranking before putting into action any remedial procedures. Common Vulnerability Scoring System (CVSS) is a tool that administrators may use to prioritize security problems according to the severity level associated with each fault. The CVSS score of vulnerability is a standard metric that is not developed for unique network architecture. Despite the fact that the frequency and impact of vulnerabilities affect the security risk level of a specific network, the CVSS score of vulnerability is a standard metric. In addition to the severity score, a number of other factors also affect the level of security risk that is posed by the organization’s underlying infrastructure. These factors include the age and frequency of vulnerabilities already present in the system, as well as the impact that exploiting vulnerability has on the system. For this reason, it is advised that, when doing risk level calculations, these components, together with the CVSS severity score, be used. This will allow for effective network security risk management.

1.3 Vulnerabilities in IoT Web Applications

The authors of [4] provide a code inspection-based strategy. To identify a number of mistakes hidden inside the process, this method makes use of code inspection. It is said that the offered approach may be used to locate each and every vulnerability in the NVD. Using this classifier might assist in more accurately identifying potential security flaws.

In addition, a web crawler was developed by Guojun and his colleagues [5]. This web spider collects papers that are connected to one another. The TF-IDF is essential to the methodology. Medeiros et al. [6] were the ones who first proposed the approach for evaluating the quality of the code. The concepts that underlie data mining are built on this methodology, which acts as the basis for those concepts. New techniques for identifying web server vulnerabilities were developed by [7].

Authors [8] have developed an innovative method for locating vulnerabilities in web applications. In addition to this, static analysis and data mining directly from the source code are used. Researchers [9] came to the conclusion that XML injection is a critical issue that exists in all web applications. The vast majority of recently published web...

Produkt-Details

• Erscheint lt. Verlag: 7.1.2025
• Sprache: englisch
• Themenwelt: Mathematik / Informatik ► Informatik ► Software Entwicklung
• Themenwelt: Technik ► Elektrotechnik / Energietechnik
• ISBN-10: 1-394-27244-8 / 1394272448
• ISBN-13: 978-1-394-27244-0 / 9781394272440