Mugdha
@scit.edu
Associates Professor
Symbiosis Centre for Information Technology
RESEARCH INTERESTS
Finance and Accounting
Scopus Publications
Scopus Publications
Mugdha Shailendra Kulkarni, S. Vijayakumar Bharathi, Arif Perdana, and Divisha Kilari
Springer Science and Business Media LLC
Abstract Understanding the complexities of buying, selling, and holding stocks is crucial for institutional and individual investors to make informed decisions. Despite their significance, many investors face challenges in this area. Accurate stock price forecasting is a vital tool that aids investors in making profitable decisions. This study evaluates stock trends and patterns with an in-depth analysis of the Bombay Stock Exchange (BSE) stock data. We utilized various techniques, including time-series analysis, machine learning, and deep-learning models. This investigation spanned two distinct datasets: one with and one without COVID-19 stock price data. By comparing the outcomes, we seek to identify the most effective model for stock price prediction. Our findings indicate that each model has its strengths and limitations. Time series models accurately forecast short-term stock prices, whereas machine learning models demonstrate superior generalization capabilities. Deep learning models, however, stand out for their ability to predict long-term stock prices more accurately. Understanding each model's performance nuances is crucial for institutional and individual investors and regulators to optimize their strategies and decision-making processes.
Rupali Kumari and Mugdha Kulkarni
Informa UK Limited
Apoorva Mohan Patil and Mugdha Shailendra Kulkarni
IEEE
Bindi Varghese, Ginu George, and Mugdha Shailendra Kulkarni
Springer Nature Singapore
Mugdha Shailendra Kulkarni, Dhanya Pramod, and Kanchan Pranay Patil
IEEE
Mugdha Shailendra Kulkarni, Kanchan Pranay Patil, and Dhanya Pramod
Informa UK Limited
Ginu George, Mugdha Kulkarni, and Bindi Varghese
Emerald
Purpose This study aims to examine the increased focus on neurodiversity in contemporary businesses. It shows how inclusive policies can capitalize on the special abilities of people with neurodiverse backgrounds, including their extraordinary problem-solving abilities, meticulous attention to detail and creative thinking. These policies benefit the individuals and contribute to a more diverse and innovative workplace. Design/methodology/approach Data was collected through semistructured interviews with HR experts and neurodivergent employees. The qualitative data were manually analyzed and coded, and themes were identified. Findings The results highlight the significant benefits of accepting neurodiversity in the workplace, enlightening the audience about its potential. For instance, artificial intelligence (AI) can be used to anonymize resumes, removing potential biases related to gender, ethnicity or age. In addition, AI can help in identifying the unique skills and strengths of neurodivergent employees, enhancing the fit between job responsibilities and their abilities. This study also emphasizes the wider effects of accepting neurodiversity on employee satisfaction, productivity and organizational innovation. This study promotes a deep learning framework that combines human-centered strategy with strategic methods to maximize the participation of neurodiverse workers and foster a more creative and dynamic corporate culture, convincing the audience of its benefits. Research limitations/implications This study is limited by its qualitative nature and relatively small sample size, comprising 15 HR professionals and 20 neurodivergent employees, which restricts generalizability. The sensitive nature of neurodiversity also made participant recruitment challenging, with some individuals hesitant to disclose their condition. In addition, companies were reluctant to share internal AI practices due to confidentiality concerns. The research focused on a select set of organizations, primarily from specific regions, limiting cross-cultural applicability. Furthermore, the absence of AI developers in the sample means insights into technical tool design and implementation remain unexplored, suggesting a gap for future multidisciplinary research. Practical implications This study provides actionable insights for HR professionals and organizational leaders aiming to improve neurodiverse hiring and support systems. It identifies specific AI tools such as Grammarly, Otter.ai and Pymetrics, that can be integrated into recruitment and workplace settings to enhance communication, reduce sensory overload and match roles to individual strengths. Organizations can use the deep learning framework proposed to design more inclusive policies and infrastructure. Training managers and customizing AI-driven accommodations can improve retention, engagement and performance among neurodiverse talent. This research supports firms in developing more equitable, adaptive and innovative environments aligned with diversity and inclusion goals. Social implications This study promotes a societal shift in how neurodivergent individuals are perceived and supported in the workforce. By emphasizing ability over deficit and proposing inclusive AI integration, it helps reduce stigma and encourages broader acceptance of cognitive diversity. The findings advocate for universal accommodations that do not require self-disclosure, promoting dignity and equity. Improved employment outcomes for neurodiverse individuals contribute to economic inclusion, reduce unemployment rates and challenge ableist norms. The research also aligns with broader Diversity Equity and Inclusion (DEI) movements, inspiring organizations and policymakers to build socially responsible frameworks that reflect the value of every individual, regardless of neurological difference. Originality/value This paper offers original value by exploring the underresearched intersection of AI and neurodiversity inclusion in the workplace. It contributes novel insights through qualitative analysis of HR professionals and neurodivergent employees, highlighting the role of AI in reducing hiring bias, customizing work environments and enhancing employee well-being. By proposing a deep learning framework and cataloging AI tools matched to neurodiverse conditions, this study bridges theory and practice. It uniquely positions AI as both a technological and ethical enabler for inclusive employment, making it highly relevant for scholars, practitioners and policymakers aiming to foster equitable, future-ready workplaces.
Kanchan Pranay Patil and Mugdha Shailendra Kulkarni
CRC Press
Priyanshi Sharma and Mugdha Shailendra Kulkarni
IEEE
In the changing era of software development, Artificial Intelligence (AI) integration has become a disruptive trend, especially in the areas of software development. This paper focused on addressing existing gaps in understanding the impact of GenAI in DevOps CI/CD pipelines, the goal of this research is to fulfill the gap by investigating the potential of GenAI and evaluating the broader impact of AI techniques on CI/CD, in order to improve the effectiveness, speed, and quality of process related to development of software. Continuous Integration ensures the instant integration of any changes made to code into a shared repository whereas Continuous Delivery speeds up the delivery lifecycle, minimizes manual intervention, and automates the deployment of software to production environment.The methodology is to identify and analyze the current landscape of AI-powered DevOps tools, further investigate different AI techniques which can be applied to CI/CD pipelines, and then evaluating the impact of AI on CI/CD efficiency and effectiveness. The research use both qualitative and quantitative approaches, by a questionnaire-based survey to collect data from industry professionals.The findings shows that even if CI/CD pipelines have advancement, problems still exist, requiring the use of GenAI to achieve more benefits. GenAI emerges as a shifting technology, parallelly responsible AI is needed to address ethical concerns for AI decision-making. This research will be beneficial for software developers, DevOps teams, and organizations who wants to enhance the reliability, and ethical dimensions of their CI/CD processes. By addressing existing gaps and understanding the potential of AI in DevOps, this research contributes to both developments in theory and real-world applications in the rapidly evolving field of software development.
Mugdha S. Kulkarni and Kanchan Patil
Springer Nature Switzerland
Shivani Arya and Mugdha Shailendra Kulkarni
IEEE
As Agile project management practices continue to evolve, the adoption and effective utilization of agile tools, such as ClickUp, Rally, and Asana, have become increasingly vital. The surge in Agile adoption highlights the critical need to comprehensively examine the tools driving Agile project management. Despite the acknowledged benefits of Agile methodologies, challenges in adopting and effectively utilizing Agile project management tools pose potential bottlenecks. This study investigates the factors influencing the adoption and acceptance of Agile project management tools. The literature review exposes a deficiency in understanding the features of existing agile project management tools. There is a notable gap in the comprehension of available tools and their functionalities. To bridge this gap, an Integrative Literature Review framework is used, utilizing qualitative data analysis through NVivo software version 14. The discussion provides a detailed comparative analysis of these tools, emphasizing the importance of aligning tool features with project requirements. The research contributes to the ongoing evolution of the Agile tool landscape, offering a nuanced understanding of the factors influencing the adoption and acceptance of tools. This research equips practitioners and project managers with a deeper understanding of Agile project management tools, emphasizing the necessity for alignment with project requirements to maximize their effectiveness in diverse project contexts.
Mugdha Shailendra Kulkarni, Dhanya Pramod, and Kanchan Pranay Patil
Springer Nature Switzerland
Kanchan Patil, Dhanya Pramod, and Mugdha Kulkarni
IEEE
This study aims to investigate the complex interactions between financial decision-making and the critical elements affecting the quality of information generated by Generative Conversational AI agents. Through an empirical investigation of these dynamics, the study seeks to illuminate the concrete consequences of utilising such agents to improve decision-making in financial inclusion. Using an information quality framework-based model and Purposive sampling, we collected data via online questionnaires from 328 banking and financial services customers. The SmartPLS analysis for SEM (structural equation modelling) produced enlightening findings. The research concludes that the representative, contextual, intrinsic, and accessible aspects of information quality significantly impact financial decision-making. This emphasises how important a role these dimensions play in determining how effective financial decisions are. Our results also highlight the critical influence that financial inclusion has on the general effectiveness of financial decision-making. The study aims to empirically investigate the valuable applications of Generative Conversational agents in management practices. This study establishes a foundation for future efforts to utilise AI-powered conversational agents in the ever-changing financial services industry by revealing the complex relationships between information quality and financial decision results.
Kanchan Pranay Patil, Mugdha Shailendra Kulkarni, and Manoj Hudnurkar
IEEE
This study explores the applicability of integrating cutting-edge AI chatbots that resemble human-like interactions in the customer service area, drawing on the underlying ideas of technological readiness and anthropomorphism. The study uses 688 questionnaire responses using purposive sampling and structural equation modelling as part of its empirical research technique to learn more about how customers perceive and use AI-powered humanoid chatbots (AIPHC). The target respondents were those who had some prior experience of using chatbots. Using SmartPLS-SEM, the study determines whether anthropomorphism, i.e., AIPHC system-specific and customer personality-specific dimensions, can improve chatbot usage in the insurance sector. The results show that anthropomorphism positively influenced optimism and innovativeness but negatively impacted discomfort and security. Further optimism and innovativeness favourably impact AIPHC adoption. Insecurity had a significant negative impact, while discomfort was insignificant for AIPHC adoption. The results could help us better understand how technological readiness and system-specific theories like anthropomorphism can be used in sectors like insurance. The results highlight chatbots’ potential to raise the quality of service, simplify processes, and enhance customers’ overall experiences in the insurance sector.
Vidyavati Hemantkumar Ramteke, Shubhangi Sanghi, and Mugdha Shailendra Kulkarni
IGI Global
The increasing integration of chatbots in the travel, tourism, and hospitality (TTH) industry has transformed customer interactions, offered personalized assistance and streamlined services. Leveraging a systematic literature review and in-depth analysis, this research paper explores the security landscape of chatbots in the TTH sector, identifying potential threats and proposing targeted countermeasures for safeguarding customer data. The study outlines potential attacks on chatbot modules, highlighting risks to user privacy and system integrity. Proposed countermeasures extend beyond existing literature, addressing specific vulnerabilities identified through in-depth analysis. The paper contributes to understanding AI-specific security threats, raises ethical concerns, and identifies potential barriers to chatbot adoption. This research comprehensively analyses security vulnerabilities in TTH chatbots, offering tailored countermeasures.
Ginu George, Bindi Varghese, and Mugdha Kulkarni
Springer Nature Singapore
Mugdha S. Kulkarni, Bindi Varghese, and Ginu George
Springer Nature Singapore
Kanchan Pranay Patil, Mugdha Shailendra Kulkarni, and Manoj Hudnurkar
Emerald
PurposeThis study aims to explore the potential of artificial intelligence with AI-powered humanoid Chatbots (AIPHC) as transformative tools to improve customer service quality in the insurance sector. The usability and efficiency of integrating advanced AI chatbots that can replicate human-like interactions in insurance services will be examined by taking into consideration customers’ technological readiness and chatbots’ anthropomorphism.Design/methodology/approachThis empirical study analysed 688 customer responses collected through purposive sampling using structural equation modelling. With the help of SmartPLS 4.0, the study determines whether anthropomorphism, that is AIPHC system-specific and customer personality-specific dimensions, can influence the acceptance of AIPHC in the insurance sector.FindingsThe results show that the chatbot’s anthropomorphism positively influenced customers’ optimism and innovativeness but negatively impacted discomfort and security. Further optimism and innovativeness favourably impact AIPHC adoption. Insecurity had a significant negative impact, while discomfort was insignificant for AIPHC adoption.Research limitations/implicationsThe study determines how people will react to AI-powered information systems. The results could help us better understand how the technological readiness of customers can be used in emphasizing the significance of system-specific theories like anthropomorphism in sectors like insurance, where customer interactions and delivery of quality services are important.Practical implicationsThe results highlight chatbots’ potential to raise the quality of service, simplify processes and enhance customers’ overall experiences in the insurance sector. This study contributes to the continuing discussion on using AI technologies in customer service by considering the interplay between technology readiness and anthropomorphism. It also provides insightful information for insurance professionals and technology developers.Social implicationsAnthropomorphic humanoid chatbots can increase the availability, affordability and accessibility of essential services. They have the potential to increase users’ competence, autonomy and—possibly counterintuitively social relatedness.Originality/valueThis empirical research explores the link between anthropomorphism and technology readiness to enhance service quality provided by AI powered chatbots in the insurance sector.
Nisha TN and Mugdha Shailendra Kulkarni
Emerald
Purpose The purpose of the study is to confirm the fact that in informations security, the human factor was considered as a key carrier of the majority of attacks that an information system faces. Banking and other financial services are always top among the most attractive targets for cyber attackers. Blind phishing or spear phishing is still one of the major contributors to all malicious activities in the e-banking sector. All the counter mechanisms, therefore, revolve around the concept of how security-aware the customers are. To fool these mechanisms, attacks are becoming smarter and are searching for methods where the human involvement is diminishing to zero. Zero click attacks are one big leap that attackers are taking that removes the requirement of human involvement in initiating attacks and are moving toward an era of unassisted attacks. Even though the standard procedure and protocols are built into the banking system, they fail to detect this attack resulting in significant losses. Design/methodology/approach This paper follows a conceptual review of the upcoming concept in security and its implication in e-banking sector. The methodology adopted in this paper uses review papers, articles and white papers to conclude a theoretical model. A detailed analysis of unassisted attacks is considered from 2010 onwards till 2022. Findings This research deliberates on the methodologies of zero click attacks and gives a detailed analysis of attack vectors and their exploits. This research also identifies the likely attacks on e-banking that these vulnerabilities can trigger. Originality/value The key contribution is toward the early detection of zero click attacks, suggesting countermeasure, reducing the likelihood of these attacks and the financial impact.
Manoj Hudnurkar and Mugdha Kulkarni
IEEE
The study aims to create a method to determine the credit rating and evaluate the risks related to the Indian banking business. To create predictive models for credit ratings, Artificial Neural Network (ANN) algorithms are used, which include Multilayer Perceptrons (MLPs), Convolutional Neural Networks (CNNs), and Long Short-Term Memory (LSTM) networks. For this purpose, secondary data from over 254 Indian banks covering 2014–15 to 2021–22 were considered. The study evaluated crucial factors to predict credit scores while considering business, financial, and industrial hazards. The data includes current liability and Revenues, COGS, SG&A, R&D, EBIT, EBITDA, Interest expense, Abnormal Gains/Losses, Income Taxes, Net Income from Discontinued Op., Net Profit, and Dividends. Current and quick ratios. GDP per capita, The research's main contribution is its capacity to forecast how credit ratings for different Indian banks may change over time based on their financial, commercial, and industrial risk profiles. This cutting-edge method of credit rating prediction improves forecasting accuracy. It boosts the effectiveness of rating systems, providing invaluable insights for the banking sector to navigate these uncharted times while analysing credit risks successfully.
Mugdha S Kulkarni, Dudhia Hard Ashit, and Chauhan Nency Chetan
IEEE
Organizations strive to secure their valuable data and minimise potential damages, recognising that critical operations are susceptible to attacks. This research paper seeks to elucidate the concept of proactive cyber threat hunting. The proposed framework is to help organisations check their preparedness against upcoming threats and their probable mitigation plan. While traditional threat detection methods have been implemented, they often need to address the evolving landscape of advanced cyber threats. Organisations must adopt proactive threat-hunting strategies to safeguard business operations and identify and mitigate unknown or undetected network threats. This research proposes a conceptual model based on a review of the literature. The proposed framework will help the organisation recover from the attack. As the recovery time is less, the financial loss for the company will also be reduced. Also, the attacker might need more time to gather data, so there will be less stealing of confidential information. Cybersecurity companies use proactive cyber defence strategies to reduce an attacker's time on the network. The different frameworks used are SANS, MITRE, Hunting ELK, Logstash, Digital Kill Chain, Model in Diamonds, and NIST Framework for Cybersecurity, which proposes a proactive approach. It is beneficial for the defensive security team to assess their capabilities to defend against Advanced Threats Persistent (ATP) and a wide range of attack vectors.
Mugdha Shailendra Kulkarni, Hrishikesh Laxman Naik, and S. Vijayakumar Bharathi
IEEE
The General Data Protection Regulation (GDPR) has far-reaching implications that extend beyond the European Union (EU) borders. Its visionary concepts have worked as a catalyst, inspiring states and regions worldwide to rethink their data privacy policies. The GDPR is the foundation of an international privacy framework that crosses political and cultural boundaries, encouraging a collaborative effort to protect personal data. This study aims to explore the GDPR's multifaceted effects on data privacy, including organisational, legal, and technological aspects. To provide comprehensive insights, data was collected from the privacy policies of the top 50 websites for both the pre-GDPR and postGDPR eras. Textual analysis of these policies allowed for a comparison using keyword frequency, personal information disclosure, GDPR-related terms, and the enumeration of user rights being evaluated. To examine the structural changes in privacy rules, we use the average frequency with which a specific group of words. The finding reveals that the average word count of privacy rules has increased significantly, rising from 2355.74 to 5882.82. This shift indicates a positive trend towards increasing transparency and clarity in communicating privacy policies. Privacy policies are now written in more precise, accessible language, emphasising the essence of GDPR's objective and improving user comprehension. In the post-GDPR, key terms such as “Consent,” “Data Breach,” “Data Subject,” “Data Controller,” and “Data Protection” feature prominently, indicating more awareness and more substantial compliance obligations. Implementing GDPR user rights under amended rules demonstrates a noteworthy dedication to user data protection.
Lydia Fernandes, Mugdha Kulkarni, and Mandaar B. Pande
IEEE
Quantum Computing is a dynamic field which has evolved since its inception around 3 decades back, to the development of various powerful algorithms including Quantum Approximate Optimization Algorithm (QAOA) and Variational Quantum Eigensolver (VQE) to name a few. Today's quantum computers are not as powerful as the current breed of supercomputers. However, the existing breed of NISQ (Noisy Intermediate Scale Quantum) quantum computers have a significant potential to provide faster solutions to problems in various domains which are not just relevant for the present but also for the future such as Quantum Machine Learning in the Fintech domain. Portfolio Optimization (PO) is one such problem in research and the financial services industry at large. In principle, Quantum Computing, makes it possible to arrive at an optimal portfolio composition for a high return and low risk investment much faster than any existing supercomputer today. This review paper examines literature on classical and quantum machine learning approaches for Mutual Fund PO, analyzing 44 papers from 2003 to 2023. PO being a vast topic, we limit our scope to Mutual Funds and its driver that is the equity markets. We provide an overview to the types of problems, preferred approaches, their benchmarks, deduced conclusions, and research gaps as a comprehensive survey for diverse readers.
Mugdha Kulkarni, Juhi Tikyani, and Krishna Kumar Singh
Springer Nature Singapore