Darpan Anand
@cuchd.in
Professor
Chandigarh University
Prof. (Dr.) Darpan Anand is a Professor in the Computer Science Engineering Department at Chandigarh University, India with more than 17 years of experience in teaching, industry, and research. He is currently a member of the Board of Studies, a Member of the research Degree Committee, Outcome Based Education Coordinator, ABET Accreditation Coordinator, Research Coordinator, and the Coordinator of Projects in the Department of Computer Science, Chandigarh University. He teaches Computer Networks, Operating Systems, Cryptography & Network Security, Machine Learning, Robotics Process Automation, etc. His research interests include Information Security, e-governance, machine learning, Intelligent Information Processing, Evolutionary Algorithms, etc. He has guided several Ph.D. and PG Dissertations. He is an author/co-author of more than 50+research papers (indexed in SCI, ESCI, Scopus, etc.), 1 Textbook, 6 book chapters (IET, Springer, and Elsevier), 4 patents, SWAYAM MOOC courses,
EDUCATION
B.Tech., M.Tech., Ph.D.
RESEARCH INTERESTS
Information Security, Computer Networks, Machine Learning, Blockchain
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Vivek Bhardwaj, Shveta Yadav, Navjeet Kaur, and Darpan Anand
Springer Science and Business Media LLC
Gurpreet Singh, Puneet Kumar, and Darpan Anand
Springer Science and Business Media LLC
Amit Yadav, Munish Khanna, and Darpan Anand
Springer Science and Business Media LLC
Gurpreet Singh, Puneet Kumar, and Darpan Anand
World Scientific Pub Co Pte Ltd
Musculoskeletal disorders can significantly disrupt the quality of life of an individual. Early diagnosis and management are critical for minimizing the long-term impact of these disorders. The musculoskeletal upper extremities radiographs (MURA) and lower extremity radiographs (LERA) datasets have emerged as a significant resource in the field of musculoskeletal imaging, facilitating the development and evaluation of various AI models for diagnostic and predictive tasks. The systematic review is carried out in this work to scrutinize existing deep learning models, unraveling their capabilities in decoding complex abnormalities within the musculoskeletal system. By intricately analyzing these models, this study contributes valuable insights into their strengths, weaknesses and performance that are pivotal for advancing the field of musculoskeletal health, ultimately facilitating more informed and precise diagnostic practices. The various limitations and gaps identified during this review will undoubtedly prove valuable for researchers in this field, offering insights to increase the efficiency and effectiveness of their models for the classification and detection of abnormalities in the musculoskeletal system. This undertaking is not merely an aggregation of disparate studies but a deliberate effort to contribute to the ongoing discourse within healthcare, fostering a foundation for evidence-based practice and policy development.
Jayant Singh Jhala, Chandani Joshi, and Darpan Anand
IEEE
The rapid development of self-driving vehicles necessitates integrating a sophisticated sensing system to address various obstacles posed by road traffic efficiently. While several datasets support object detection in autonomous vehicles, evaluating their suitability for different weather conditions globally is crucial. In this study, we present deep learning models trained on a novel dataset derived from YouTube videos recorded from Indian car’s dashcams. These videos capture a wide range of conditions, including rain, fog, daytime, hazy and night-time driving scenarios prevalent in India. The dataset comprises a total of 1450 annotated images depicting vehicles and other road assets across six different classes. In this work, performance analysis of the YOLOv8 models trained using an existing dataset was compared with the model trained on an expanded version using the proposed weather-specific dataset. The results demonstrate improved accuracy metrics of 91.3%, 84.5%, and 91.2% for Precision, Recall, and mean Average Precision (mAP) upon integrating the proposed dataset. The model trained on this diverse dataset exhibits heightened robustness, proving highly beneficial for autonomous and conventional vehicle operations in India’s dynamic traffic environments. This research contributes to advancing object detection capabilities crucial for autonomous driving technologies in real-world settings.
Vishan Kumar Gupta, Paras Jain, Darpan Anand, Gamini Dhiman, Sarvesh Vishwakarma, and Anurag Aeron
IEEE
According to the banking supervision and regulation of the worldwide, credit risk approximation of a bank/financial institute is not only confined within a product line or value chain anymore. Today’s banking regulation is putting pressure on the banks to deliver the risk measure values by regarding the risk considerations more broadly than just product and trading desk level together with the changes in the market environment and customers’ commitments. Hence, it is deprived of fixed parameters and wholly relies on different parameters and huge data documents. Given very fast progress in the field of data availability, generation, storage, and data processing, machine learning now occupies a strategic position in all areas of the banking business and technologies. The usage of machine learning is extremely appreciable in the construction of credit risk modelling and loss estimations. Risk practitioners are inclining their focus toward applying the advantage of deep learning for the nonlinear relationships of substantive variables and default risk.
Nilesh N. Maltare, Safvan Vahora, Surendra Kumar Shukla, Ghanshyam Raghuwanshi, Sarvesh Vishwakarma, Darpan Anand, Vishan Kumar Gupta, and Jaishree Meena
IEEE
The best software development approaches are encapsulated in design patterns. Parallel design patterns are the patterns used in the development of parallel software. The proposed mathematical representation of design patterns provides better representation and understanding. The hybrid representation of the pattern includes all the key features such as visualization, relationship, class representation, and order specification from the existing languages (i.e., LePUS, ExLePUS, BPSL, and LOTOS) of the design patterns specifically for the parallel design. Furthermore, this proposed parallel design representation incorporates ordering constructs such as sequencing, synchronization, and parallel processing as a unique approach. The proposed pattern representation helps with the structural, behavioral, and collaborative aspects of the parallel design. The mathematical and visual representation of this approach leads to the robust and comprehensive parallel design of the software.
Richard Essah, Darpan Anand, and Abhishek Kumar
Springer Nature Singapore
Amit Malik, Edeh Michael Onyema, Surjeet Dalal, Umesh Kumar Lilhore, Darpan Anand, Ashish Sharma, and Sarita Simaiya
Elsevier BV
Ruchi Doshi, Kamal Kant Hiran, Maad M. Mijwil, and Darpan Anand
IGI Global
Business intelligence (BI) is no longer adequate to handle the day-to-day operations of any firm considering the ever-increasing volume of data and the resulting overload. As the amount of data grows, it becomes increasingly difficult to evaluate, making the introduction of a decision-making methodology that can be described as real-time BI, very taxing and cumbersome. Because of this, it is becoming increasingly difficult to implement effective decision-making at the enterprise level that was driven by BI, so that the company may remain robust and resilient to both man-made dangers and natural calamities. With today's sophisticated malware and the growing importance of the Internet of Things (IoT), we require a more sophisticated intelligence system, which we currently refer to as Artificial Intelligence (AI). We have a better chance of surviving a cyber-attack thanks to AI and its two other subsets, Machine Learning (ML) and Deep Learning (DL). These technologies strengthen our organization's day-to-day operations and help us make more reliable decisions as stakeholders.
Vishan Kumar Gupta, Sarvesh Vishwakarma, Darpan Anand, Arun Kumar, Harish Tiwari, and Ankit Kumar
IEEE
To know whether specific chemical compounds can interfere with bodily functions that might have a negative impact on health, scientists have developed an effective bagging-based ensemble model for better activity detection. Here, authors are utilizing a variety of physicochemical characteristics (molecular descriptors), where an in-silico method has been developed for the detection of the activity of estrogen receptor alpha ligand binding domain pharmaceuticals. To extract features, one uses the PaDEL-Descriptor. The 7741 drug molecules in the ER-LBD dataset have a total of 1444 features, with 438 of them being active and 7303 of them being inactive. This dataset has a lot of features and is very unbalanced. First, a problem with class imbalance is fixed by Smote algorithm and selecting important features with the CFS algorithm. An ensemble learning-based prediction model is created, where categorization was done using four existing classifiers i.e., SVM, decision tree, AdaBoost, and random forest. With the help of this strategy, the proposed bagging- based detection model provides outperformed accuracy. To assess the model's consistency across all target classes, the K-fold cross validation is used.
Harsh Bansal, Divya Gupta, and Darpan Anand
IEEE
As the Fourth Industrial Revolution marks the beginning of an era of digitization, key principles like inter-operability, decentralisation, and virtualisation are significantly impacting a variety of fields, including education. Aiming to combine traditional educational techniques with cutting-edge ICT-enabled tools, including artificial intelligence, automation, and blockchain, this paradigm coined as “Education 4.0” intertwines with “Industry 4.0”. With its intrinsic qualities of decentralisation, immutability, consensus, and transparency, blockchain stands up as an effective solution to problems like governance and credential credibility of the courses in which the student is enrolled can be resolved. This paper has contribution in four folds. This paper makes four significant contributions to literature. It begins by highlighting the goals of Education 4.0 and the relationship between modern teaching methods with cutting-edge technology. It then delves into an analysis of blockchain protocols, spotlighting their alignment with the DeFi protocol stack. The next part divides blockchain projects into four distinct categories, following which reviews diverse blockchain-powered projects in education, evaluating their implementations, advantages, and potential drawbacks.
Kaljot Sharma and Darpan Anand
CRC Press
Amandeep Kaur, Rakesh Sahu, and Darpan Anand
IEEE
This study established a novel routing technique that is dubbed reinforcement learning for SIoT (MCTAR-SIoT). Its purpose is to ensure that data interchange in IoT networks is both energy-efficient and safe. At the subsequent hop node selection, this technique used three measures: Communication Trust (CT), Energy Trust (ET), and Hop Count (HC). The CT assures that nodes can be trusted, the ET ensures that nodes have adequate energy, and the HC provides a route with reduced time. By integrating all of these indicators and giving each characteristic its own weight, we are able to develop a new CRM metric. The first feature ensures that communications are kept private, the second extends the device's operational lifespan, and the third shortens the response time. Each node that is picked is the next-hop node, and it is responsible for transmitting the data to the destination node via it. The CRM uses deep reinforcement learning to choose which values to use. Extensive simulations were run for experimental validation, and the performance of the network was evaluated using MDR. These simulations included altering the proportion of members of the network who were malevolent. They saw an increase in throughput of 3% from CNN and 7% from SVM, respectively. Next, they saw an improvement in their MDR of roughly 2.6341% from CNN and 7.8112% from SVM, respectively. In conclusion, the NL is enhanced by 8% from OCNN and 10% from SVM, respectively, resulting in an improved score.
Amandeep Kaur, Rakesh Sahu, and Darpan Anand
IEEE
A recent trend in academic study is the Social Internet of Things (SIoT), which integrates social network theory into the various layers of the IoTs to open up novel avenues for the IoTs' future growth. Slot seems to be a subset of lots that utilizes intelligent equipment and humans as nodes, social networks as the organization type, the social connection between things, things and humans, and humans, and formats study models and techniques with social media network characteristics in order to realize the connection, service, and usage of the lots. In addition, Slot is a realization of the innovation, architectural style, and application of lots using social network research methodology, and it helps promote the incorporation between the real and the virtual cyberspace, contributes to the realization of the lots, broadens the study scope of social networking, and offers a new remedy for the particular problems of the lots. Researchers must, therefore, have a firm background in the SloT developments in machine learning. SIoT uses Machine Learning (ML) methods to determine the features of connected devices. This special issue solicits scientific research articles that provide a picture of the current state of SIoT research. In this research improve the features of SIoT by residual network mapping and learning by XGBOOST
Jayant Singh Jhala and Darpan Anand
IEEE
The Indian automobile industry had its highest-ever annual domestic passenger vehicle sales last year. A total of 3.793 million or 37.93 lakh units were sold in the country in 2022, which is 23.1 percent higher than the preceding year. Similarly, the used car sale is also be increased day by day. The actual and reasonable rates of used cars are important to sale and purchase so that, buyers and sellers will be get benefited. The disparity in prices due to various characteristics or features consistently making prediction of price a difficult job. In the matter of used car price prediction, it has been equally tough. It is challenging to determine when the advertised price is indeed legitimate. Used car prices are highly affected by features like new car price, engine power (cc), maximum power (bhp). For the sake of predicting used car prices on ground of its characteristics, this research target to generate machine learning models incorporating multiple linear regression, decision tree regression and random forest regression. The Car Dekho data set that is used, initially had 13 attributes and 19974 records. In this Research Field, significant study has been carried out; although, not all of them utilized Scikit-learn. Our Suggested models produced accuracy of 94.10% with random forest regression whereas 92.45% with decision tree regression and 89.85% with multiple linear regression.
Shikhar Bhardwaj, Sandeep Harit, N.A. Shilpa, and Darpan Anand
Inderscience Publishers
Richard Essah, Darpan Anand, and Surender Singh
Springer Nature Singapore
Richard Essah, Darpan Anand, and Surender Singh
Elsevier BV
Gurpreet Singh, Darpan Anand, Woong Cho, Gyanendra Prasad Joshi, and Kwang Chul Son
MDPI AG
The practice of Deep Convolution neural networks in the field of medicine has congregated immense success and significance in present situations. Previously, researchers have developed numerous models for detecting abnormalities in musculoskeletal radiographs of upper extremities, but did not succeed in achieving respectable accuracy in the case of finger radiographs. A novel deep neural network-based hybrid architecture named ComDNet-512 is proposed in this paper to efficiently detect the bone abnormalities in the musculoskeletal radiograph of a patient. ComDNet-512 comprises a three-phase pipeline structure: compression, training of the dense neural network, and progressive resizing. The ComDNet-512 hybrid model is trained with finger radiographs samples to make a binary prediction, i.e., normal or abnormal bones. The proposed model showed phenomenon outcomes when cross-validated on the testing samples of arthritis patients and gives many superior results when compared with state-of-the-art practices. The model is able to achieve an area under the ROC curve (AUC) equal to 0.894 (sensitivity = 0.941 and specificity = 0.847). The Precision, Recall, F1 Score, and Kappa values, recorded as 0.86, 0.94, 0.89, and 0.78, respectively, are better than any of the previous models’. With an increasing appearance of enormous cases of musculoskeletal conditions in people, deep learning-based computational solutions can play a big role in performing automated detections in the future.
Darpan Anand and Aashish Kumar
Wiley
Shardeep Kaur Sooch, Darpan Anand, and Rajesh Deorari
IEEE
Early brain tumor detection has become important to provide timely diagnosis and treatment. Several methodologies are focusing to minimize the manual efforts required for diag-nosing by increasing not only the accuracy but also the speed of detection. This proposed methodology includes Otsu's threshold-based segmentation technique after which feature extraction is done by Grey Level Co-Occurrence Matrix (GLCM) to extract 13 intensities based and textual based features. The classification is done through the hybrid model of K-Nearest neighbor and Random Forest. The final outcome is generated by majority voting which castes its vote to either one of the above hybrid models. The results are compared to existing algorithms through various performance parameters which includes how accurate the result is, recall, specificity and time taken for execution
Talib ul Haq, Darpan Anand, and Nitika Kapoor
IEEE
Peer reviewing plays an important role if a re-searcher wants to publish his/her manuscript. Peer review is the impartial evaluation of your research article by subject matter experts in your area. Its objective is to assess the manuscript's quality and publishing readiness. This paper will be considering these peer reviews and use computational linguistics to relate them with acceptance of the manuscript. This paper will focus on text conversions in depth and discuss the flow of natural language processing. Natural language processing is a field of computer science and artificial intelligence that studies how computers interact with human languages, In specific it is an automated process of analyzing human language in order to gain information from it. Automation is the ultimate goal for efficiency driven organisations and Natural language processing is playing an important role in it. A machine can fully understand humans only when it communicates it with human language, its like communicating to someone who speaks your native language versus someone who speaks language you hardly can speak or understand. Natural language processing removes this gap. Text processing and text summarization, Automatic translation of languages, user interfaces, speech recognition and expert systems are examples of natural language processing applications. Chal-lenges for processing the complex language system of humans are far too many as human emotions and expressions through voice and body language are not fully decode in data. In near future everyone will have to communicate with a machine for household work as well as space work, so it becomes important to understand the working of Artificial Intelligence and Natural language processing is an important part of it.
Richard Essah, Darpan Anand, and Surender Singh
IEEE
Information technology changes the world and leads to the transformation of the industry to Industry 4.0. Further, this development includes the part of artificial intelligence, which will become the industrial revolution 5.0. This transformation impacts all the associated fields of quality human life. Agriculture is also a very important field for bettering human life. There are various types agriculture technology and this technology includes the Internet of Things, artificial intelligence, machine learning, and many more. The Internet of Things is the technology used worldwide to connect all the machines. For monitoring the various agriculture task, sensors are on board for communication through a low-power wireless sensor network to promote sustainable solution. Various IoT and other information and communication technology- based applications in the field of agriculture are reported in the literature. This paper aims to highlight the major contributions of information technology, information communication technology, the Internet of Things, artificial intelligence, machine learning, and many more in the field of agriculture. These technologies are used to provide solutions and transformations of various agriculture mechanisms and processes, including farm management systems, crop yield projections, crop monitoring, and smart agriculture. At ease selling and buying of agricultural yields to interested parties, Crop disease control, Crop monitoring, Crop production cycle enhancement, Soil testing method and Smart Agriculture services. Finally, the manuscript tries to emphasize the importance of implementing some critical processes for crop yield procurement through the technology related to computer science. The problem is addressed in the context of the Cocoa crop in the specific region of Africa, i.e., the Country Ghana. The purpose of the study is to review crop testing based on IoT and machine learning.