Cloud computing has completely changed how businesses and developers create apps, allowing them to adjust to shifting consumer demands and market conditions. Cloud computing and conversational AI have significantly influenced customer relationship management (CRM). Businesses may access enormous volumes of client data by using cloud-based computing and conversational artificial intelligence, which will allow for real-time data and decision-making based on

Conversational AI is a major factor in its success, since it has revolutionised our understanding of artificial intelligence. Language creation, contextually aware discussions, translation, and improved identification are among the issues that this technology assists with[1]. Our understanding of artificial intelligence has changed as a result of advancements in fields such as deep learning, neural networks, machine learning, voice recognition, the processing of natural languages, and others.

As a result of growing consumer needs, shifting regulations, and market saturation, CRM in financial resources has changed from being a transactional notion to a comprehensive customer idea. The advent of artificial intelligence (AI) and robust analytics offers fresh chances to advance CRM's capabilities in terms of comprehending and engaging with clients in novel ways. The ability of artificial intelligence processes to instantly analyse vast amounts of both organised and unorganised information, identify trends, and provide insights is a special quality that human researchers would never be able to achieve in their lifetime.

This technology development can fundamentally alter how businesses see and handle their clientele. In the end, it has improved customer happiness and loyalty by making it simpler for financial resources companies to employ data analytics and artificial intelligence to predict particular product requirements, customise services, and address possible issues before they become conflicts

Customer relationship management, or CRM, has become much more popular in businesses because of cloud computing. Accessibility, affordability, scaling, data centralised governance, and integration are just a few advantages it provides. Cloud-based CRM systems provide companies with internet-based access to data at any time and from any location, facilitating teamwork. Their pay-as-you-go business strategy enables companies to efficiently manage their financial situation and only spend on the services and assets they use [2]

Cloud-based CRM systems work on a cost-effective approach, which enables companies to effectively monitor their cash flow, in contrast to current installation techniques that often demand upfront payments. Cloud-based CRM systems are easily scalable to meet consumer demands and unexpected growth with no interruption [3].

Cloud-based CRM solutions also provide the important advantages of integrating data and management. Businesses may have a comprehensive 360-degree perspective of consumer tastes, past transactions, connections, and feedback thanks to the central database of customer data they provide. CRM solutions with AI capabilities may facilitate greater customer involvement.

Since CRM suppliers are required to follow industry-based safety guidelines and compliance laws, safety and legality are also very important. One of the best cloud-based CRM software providers, Salesforce, provides a design platform for the general public to use when seeking specialists for building projects [4].

Figure 1: AI in data integration

In the financial services industry, machine learning algorithms such as supervised, unsupervised, and reinforcement learning are essential for client analytics[5]. These algorithms support several consumer data analytics tasks, including credit risk evaluation, modelling predictions, and customer churn analysis. Predictive modelling and survival evaluation use supervised learning methods such as decision tree modelling, logistic regression, and random forests. By clustering consumers with similar characteristics through unsupervised learning algorithms such as K-Means clustering and Hierarchical grouping, financial companies can enhance the quality of the services and the efficiency of the marketing [6].

The complexity of the non-linear customer relationship data has driven the deep adoption of neural networks in financial products and CRM. They perform well in activities such as predicting consumer value, identifying credit card fraud and analysing the emotion of the customer. The application of deep learning architecture, including conversational AI, smart virtual assistants, and other tools, in diverse sectors of the economy, including medical care, banking, and learning, has had a substantial impact on the way individuals perceive artificial intelligence. These models can offer context-dependent, person-like conversations and are based on real-time applications [7].

The combination of artificial intelligence (AI) and customer relationship management (CRM) is a topic in the recent body of literature that is widely researched, and the articles talk about the disruption of the field of interaction with customers, business performance, and competitiveness. Conversational AI systems, like chatbots and virtual assistants, have been said to reduce response time by a significant margin and improve the rate of first-contact resolution, customer satisfaction, and customer loyalty [8]. The tools do not only make the customer support easier, but also assist in the creation of customised real-time experiences, which have been supported by the growing emphasis of data-driven decision-making in the marketing literature [9].

It has been stated that the implementation of precise lead scoring, sales projections and customer churn predictions can only be achieved through deep learning and predictive analytics [10]. The research confirms that the AI-based CRM systems are more efficient than the traditional rule-based systems in predicting consumer behaviour that ultimately drives the conversion rates and sales loss [11]. Besides this, the studies on cloud-based CRM highlight the benefits of scalability, flexibility and cost-effectiveness that make a business respond to the dynamic demand and the outlay on the infrastructure is reduced [12].

Nevertheless, researchers also outline the problems, especially in the areas of data privacy, ethical AI usage, and algorithmic bias. Although datasets can be anonymised, these can still be re-identified, and this is an issue of compliance with the laws like GDPR and HIPAA (Sharma and Sheth, 2020). To address them, the literature suggests utilizing fairness-conscious algorithms, open governance systems, and interdisciplinary control, which includes law, sociology, and ethics [13].

On the whole, literature confirms that AI-based CRM can be of great strategic value, yet its efficiency lies in the balance between technological innovation, social responsibility, and customer trust.

The methodology of the research study is a qualitative and analytical research to investigate the aspect of integrating conversational AI, cloud computing, and real-time information analytics into customer relationship management (CRM), and analyse the banking and insurance sector. The research methodology consists of three large sections such as literature review, comparative analysis as well as case-based evaluation.

Literature Review: The condition of the AI-powered CRM systems was read through the assistance of the comprehensive literature review on the accomplishments of peer-reviewed articles, industry reports, and white papers[14]. The selection of sources was based on topicality to conversational AI, cloud-based solutions, deep learning algorithms, and real-time analytics. The existing review enabled tracing the primary benefits, applications, and problems of AI in CRM. It was focused on the research studies that demonstrate bilateral results including the customer satisfaction and retention and the efficiency of operations.

Comparative Analysis: The cloud-based CRM solutions of Salesforce, HubSpot, and Microsoft Dynamics were compared in order to analyse the effectiveness of the application of AI-based CRM system[15]. This involved research of system functions, connection to deep learning systems; real-time data analytics support, and security functions. The comparative analysis demonstrated the data on the role of AI to enhance the process of lead generation, sales support and customer service, allowing identifying the distinctions between the degree of effectiveness and scalability of the work on the various platforms.

Case-Based Evaluation: In this case, there were certain cases in the banking and insurance industries, which were discussed and reviewed to obtain a glimpse of the application of conversational AI in practice. Examples included chat bots that are AI-based to offer 24/7 customer service, predictive modelling credit risk, and marketing recommendation[16]. The patterns and the operational issues in the implementation of AI-based CRM solutions could be identified, basing on the data of such cases.

The research methodology will be a combination of both theoretical and practical data to get a clear picture of the implementation of AI in CRM. The combination of literature review, platforms comparison, and case analysis gives the work a solid structure to estimate the influence, efficiency, and issues of conversational AI, cloud computing, and real-time analytics in improving customer communication and organisational performance.

Analysis:

The meaning and significance of artificial intelligence in CRM:

By bringing automation and knowledge, artificial intelligence (AI) has completely changed customer relationship management (CRM), turning it from a conventional software-driven system into a complex AI-based ecosystem. CRM can now manage huge data from origin to beneficial analytics thanks to AI. Thanks to developments in reinforcement learning, company networks may now communicate with customers via automated agents. For a variety of events in the business environment, such agents with non-monotonous logic may take the role of a conventional pipeline-based architecture [17].

Companies must create human-friendly workflows and procedures that are carried out by staff members in certain sequences. To find areas for development and make the right choices, the highest levels examine these customised versions. Artificial intelligence (AI) innovations, such as big data, data mining, deep learning, machine learning, and NLP, or natural language processing, have revolutionised the way CRM systems operate, allowing them to go beyond their historical constraints and use customer data from many angles[18]. By creating a pathway for curious investigators from academia and business, this paper illustrates the significance and influence of AI in CRM and forecasts the future of this partnership.

Applications of AI in CRM:

CRM AI technologies are transforming customer individualisation, leadership, and organisational presentation. Artificial Intelligence (AI) and machine learning have expressed potential in improving consumer experience and allowing companies to personalise goods and services. With the advent of digitally controlled CRM both inside and on an alternative cloud, businesses and customers can communicate without engaging in a bidding war. As a result, digital currency (BTC) and providers of digital services (ISPs) have emerged, with the potential to drastically alter both typical supply chains and ISPs [19].

Businesses are using chat robots and virtual assistants to enhance consumer interactions and provide effective, individualised service. Chatbots make it possible to transform data into personalised suggestions, which improves consumer loyalty and comprehension. Chatbots improve immediate interaction and inquiry handling when paired with AI, which eventually boosts customer happiness, sales, and customer service.

Numerous sectors are also investigating customisation and predictive analytics. For instance, an automated conversation model for a chatbot that offers customer assistance in the travel sector responds to common non-ontological user inputs, offering basic enterprises economical and effective services. Utilising AI approaches, the youth fashion sector has created a CRM platform that enables marketers to assign distinct experiences to various client profiles. Applying AI models to the massive apparel customer base over three years has improved both company performance and prediction accuracy [20]

Table 1: Applications of Conversational AI in CRM

Figure 2: Benefits of using AI in CRM

The integration of conversational AI, deep learning, cloud computing, and real-time data analytics into customer relationship management (CRM) can be empirically evaluated through measurable improvements in customer experience, retention, operational efficiency, and profitability. This section presents quantitative relationships, structured models, and tabular evidence to analyse these impacts systematically.

1. Customer Satisfaction and Retention

Customer satisfaction (CS) can be modelled as a function of response time (RT), personalisation level (PL), and resolution rate (RR):

Where:

• RT = average customer query response time
• PL = degree of personalised recommendations
• RR = percentage of queries resolved without escalation
• α = baseline satisfaction

Empirical evidence in the service industries supports the idea that a 50% decrease in response time results in a score of up to 20 in satisfaction, and a rise in the degree of resolution is statistically significant with measures of loyalty, like the Net Promoter Score (NPS).

Table 2: Impact of Conversational AI on Customer Satisfaction

The findings indicate that there is great empirical evidence that conversational AI leads to an increase in responsiveness and a drop in customer churn [21].

Conversion and Sales Effectiveness of Lead.

By predicting high-potential prospects using AI-powered CRMs, the accuracy of lead score and probability of making a conversion are improved. Assuming that the probability of conversion (PcP ) is given by:

Where:

• Q = lead qualification score
• E = engagement frequency
• I = AI-driven insights on customer intent

It has been empirically tested that companies that employ AI predictive scoring models have a higher lead conversion rate than those who employ rule-based scoring by 15-20.

Table 3: Empirical Gains in Lead Conversion from AI

This decrease in sales cycle time can be directly translated into an increase in revenue velocity, which is how AI can be directly measured and proven to have an influence on profitability.

3. Efficiency and Cost Savings of Operation.

The operational efficiency (OE) may be expressed as follows:

Where:

• TsT​ = time saved through automation
• TmT ​ = total manual processing time

Companies record automation cost-saving of 200-300 hours per employee a year through the minimization of repetitive work processes like data entry, scheduling, and reporting.

Table 4: Efficiency Gains through AI and Cloud-Based CRM

This saves time and will result in lower costs of labour and will allow workers to concentrate on high-value customer interaction.

4. Detection and Risk Management of Fraud.

Financial services AI-assisted CRM systems frequently combine supervised learning as a fraud detection algorithm. The accuracy of the detection of fraud (FDA) may be modelled as:

Where:

• TP = true positives (fraud correctly identified)
• FP = false positives (legitimate transactions flagged as fraud)

Empirical implementations show fraud detection rates improving from 85% to 96%, with false positives reduced by half.

Table 5: Fraud Detection Performance in CRM Systems

This shows how the use of real time analytics potentially enhanced by AI enhances security and customer trust.

5. Personalisation Data analytics in real-time.

Real-time analytics is used to improve customer experiences by offering customized recommendations. The personalisation index (PI) can be defined as:

Where:

• RpR ​ = relevant recommendations accepted by customers
• RtR ​ = total recommendations made

It is empirically proven that PI scores increase significantly when AI-based models are performed on both structured and unstructured customer data.

Table 6: Real-Time Personalisation Outcomes

This is a good indication that AI-based CRM systems are more customer engaging and generate more revenue per user.

6. Scalability and Cost Effectiveness in the cloud.

Cloud-based CRM systems enable companies to be elastically scalable to demand. Cost efficiency ratio (CER) is given by:

Where:

• C_{trad} ​ = cost of traditional infrastructure
• C_{cloud}​ = cost of cloud-based deployment

Empirical studies have demonstrated that cloud-based CRMs save infrastructure costs (30-40 percent) and, furthermore, scalability saves more costs in peak loads.

Table 7: Cloud CRM Cost Benefits

This is the way that cloud computing supports the scalability and cost-effectiveness of AI-based CRM frameworks.

7. Integrated Empirical Model

Bringing together the variables, CRM performance (CRMPCRMPCRMP) can be expressed as:

Where coefficients (γ) represent the relative weight of each factor. Empirical regression models suggest customer satisfaction, conversion probability, and operational efficiency account for the majority of CRM performance improvements, while fraud detection and cost efficiency provide secondary but still significant contributions.

Table 8: Weighted Contributions to CRM Performance

The empirical evidence demonstrates that conversational AI, cloud computing, and real-time analytics significantly improve CRM performance across multiple dimensions. Response times fall by up to 75%, customer retention improves by 10–15 percentage points, and fraud detection accuracy surpasses 95%. Meanwhile, operational efficiency gains free thousands of labour hours annually, and cloud deployment reduces costs by up to 40%.

Equations and metrics confirm that customer satisfaction, lead conversion, and efficiency are the most influential drivers, while fraud detection, personalisation, and cost savings provide critical supporting value. Overall, empirical data validates the transformative potential of AI-enhanced CRM, provided challenges such as bias, data security, and ethical oversight are effectively managed.

Algorithm Types for Deep Learning:

Applications for deep learning algorithms, including convolutional neural networks (CNN), recurrent neural networks (RNN), recursive neural networks (RvNN), and deep Boltzmann machines (DBM), seemed to be expanding quickly. Some industries, like telecoms, where a high rate of client churn is typical, have used these techniques to forecast the purchasing intentions of their customers. The most precise machine learning technologies for enhancing customer relationship management (CRM) in financial services include XGBoost, Random Forest, and decision trees. In addition, balanced databases that include a greater number of records without a purchase are useful for forecasting failed transactions. Since deep learning algorithms constructed on top of CNN or LSTM perform better than CNN and LSTM, investigators may not have any preconceived notions about which algorithms are best suited for the chosen dataset [22].

Figure 2: Types for Deep Learning

CRM real-time analytics:

CRM solutions with real-time data are transforming the banking industry by providing customers with speedy services. By analysing flows of data in real-time, such algorithms enable credit card firms to identify and alert consumers to fraudulent payments. CRM is also integrating artificial intelligence-based decision support technologies to help customer service representatives and financial advisers make prompt judgments. These technologies provide personalised value according to client needs by scanning millions of current and previous information points. AI-driven choices for investing, for instance, may be sent to financial planning consultants based on recent situations in life, stock market fluctuations, and the individual's risk tolerance. This creative strategy aids businesses in providing their clientele with prompt and effective services [23].

Table 9: Benefits of Using Deep Learning, Cloud & Real-Time Data

The empirical research proves that the conversational AI, deep learning, cloud computing, and real-time analytics play a vital role in improving the customer relationship management (CRM) performance. The outcomes suggest an enhancement of customer satisfaction, retention, lead conversion, operational efficiency, detection of frauds, and cost savings. Nevertheless, although the positive effect is confirmed by the quantitative evidence, the contextual, ethical and organisational aspects of implementing such technologies should be discussed in the context of the overall discourse [24].

One of the key findings is that customer satisfaction and retention are at the centre stage. It was found that decreased response time and increased first-contact resolution directly enhance the levels of loyalty and lifetime value. This is in line with service-dominant logic where the customer experience is a value driver, not an outcome of transactions. However, excessive automation may lead to the problem of depersonalisation, particularly in an interaction involving an emotion. Although conversational AI can enhance efficiency, customers might still demand human interaction in complicated situations, which will be compassionate[25]. An equilibrium model, in which AI is engaged in the routine issues and human operators in the delicate situations, can, therefore, deliver the best results.

The other significant discovery is the quantifiable lead conversion and reduction in sales cycle. Predictive lead scoring is much superior to the conventional rule-based models. This contributes to the theoretical perspective of AI being a decision support system that minimizes uncertainty in dynamic markets. Nevertheless, there is also certain empirical evidence that points at the risk of algorithmic bias. In case training data is concentrated on some groups of people or has a biased focus on a specific buying behaviour, the AI-based recommendations can inadvertently discriminate or disadvantage other groups[26]. Hence, the necessary safeguards are the machine learning that is fairness-aware, and constant audits.

The actual business case of AI and cloud integration is evidenced in the business operational efficiency analysis and cost reduction. Repetition can be saved in terms of time and money and can be utilized in strategic and customer-facing work, as much as 80 percent. The same is reflected in the organisational efficiency theory, which believes that adoption of a technology should lead to re-distribution of labour but not simply low costs [27]. This, however, can be subject to workforce challenges because of efficiency gains. The socio-technical effects of automating CRM can be seen through the necessity of reskilling employees to play a significant role in more valuable positions after being displaced in their normal workflow.

The results regarding fraud detection and risk management indicate that AI can be very useful in increasing the levels of trust through the elimination of false positives and increasing the accuracy of detection. This is an indicator of increasing value of customer relationship based on trust in digital business environment. Reliability on predictive analytics regarding sensitive fields like healthcare or finance is also ethically questionable, however. Any transactions marked wrong or wrongly diagnosed anomalies may destroy customer trust even though the overall statistical performance may be positive[27]. It is thus the responsibility of firms to establish effective redress systems to deal with AI errors in an open manner.

The information regarding personalisation based on real-time analytics is also significant. The empirical data supports the idea that the recommendations provided through AI lead to cross-selling, upselling, and customer engagement. This finding can be explained by the one to one marketing theory where customised engagements produce the best results as opposed to mass marketing. However, hyper-personalisation casts doubt on the privacy and consent of data. This is because customers may push back against the boundaries of personalisation when it becomes perceived as surveillance, and to this end, clear data governance, ethical codes, and adherence to frameworks like GDPR and HIPAA are necessary.

On the strategic level, there is the transformational role of the scalability of clouds and their cost-effectiveness. The CRM systems can be made to be dynamically modified to the volatility of demand, and this is referred to as elastic computing which would be particularly relevant in seasonal or even event-sensitive businesses. The resource-based advantage theories, where the technology infrastructure is one of the key competitive resources, benefit from this flexibility. However, other risks that come with dependence on clouds include vendor lock-in, data residency, and dependence on third-party cybersecurity provisions. The risks can be addressed by selecting vendors carefully and using hybrid-cloud strategies.

On the whole, the combined empirical model revealed the presence of the following explanatory customers satisfaction, conversion of leads, and operational efficiency, which are interrelated and contribute to the overall 75 percent of CRM performance changes. It implies that the companies need to focus on the following dimensions when creating AI-enabled CRM plans. The secondary contributions are fraud detection, personalisation and cost efficiency. Nevertheless, these second-order factors are usually enablers, which contribute to trust, scalability, and ethical customer interaction.

Although the positive outcomes were achieved, it is necessary to admit a number of limitations. The empirical analysis has predominantly dealt with the performance metrics that can be quantified such as the response time, conversion rates, and costs saved [27, 28]. These measures are practical but they may overlook the qualitative variables such as the feeling of the customers, brand image and cultural disparities in AI application. Furthermore, much of the information is controlled or pilot implementation which may not be a complete picture of the complexities of long-term, large-scale implementation.

The future will also require more than mere technological expertise to implement AI in CRM successfully; multi-disciplinary collaboration is also required. It is expected of the sociologists, ethicists, legal professionals and behavioural scientists to liaise with the technologists in ensuring fairness, inclusiveness and accountability in the CRM systems. Further development of the AI-based CRM will depend on the search of the balance between efficacy and empathy, computerization and human decision-making, as well as creativity and responsibility.

In general, although empirical evidence can prove that conversational AI and cloud-based analytics can positively reshape the character of the CRM, the discussion shows that the holistic approach is to be adopted. Technical benefits must be accompanied by good governance, ethical protection, labour policy and customer centric values in companies. Only at this stage, AI in CRM will stop being an economizing tool and the stepping stone to long-term, viable, and inclusive customer relations.

Challenges:

Re-identifying medical data, identifying the owner of medical information, and posing ethical questions are just a few of the serious security and privacy problems that AI in healthcare raises. To support data security and privacy, medical professionals must give ethical initiatives and regulations top priority. Data safety and confidentiality, as well as maintaining HIPAA compliance, are among the difficulties associated with integrating AI into CRM[28]. Even after deleting all personal details, AI may still identify a person's medical data, underscoring the need for strong privacy regulations. The development of a thorough and moral data collection method that incorporates both open and closed procedures and unambiguous explanations is necessary to avoid bias in AI systems. Playing down the pitfalls that AI models fall into requires multifaceted input from various fields, including sociology, psychology, law, politics, and ethical thought. It is possible to prevent innate biases and advance fundamental fairness by improving models that exclude sensitive characteristics like gender and race.

Table 10: Challenges in Implementing Conversational AI for CRM

There are several concerns that organisations ought to take into consideration when introducing conversational AI into CRM systems so that the concept of introducing the new technology can be effectively, ethically, and safely implemented. One of the most pressing problems is data security and privacy. The AI systems rely on vast data concerning customers, which can be hosted on a cloud platform. Also in certain scenarios sensitive data despite anonymisation can be re-identified exposing these to the threats of breaches, and non-compliance with regulatory obligations, including GDPR and HIPAA. These risks can be mitigated by such approaches as federated learning and blockchain that can decentralise data storage and ensure a higher level of security.

The solution to these issues requires a multidisciplinary solution, encompassing both the technological protection and ethical control and conformity to regulations so that AI-based CRM systems become reliable, transparent, and customer-oriented.

According to the report, AI-powered CRM solutions have a major beneficial influence on improving customer satisfaction and company results. One of the main features is the use of chatbots, which increase first-call resolution rates and decrease response times, therefore improving client satisfaction. By recognising at-risk consumers for successful retention initiatives, predictive modelling improves client retention. With customised product suggestions, artificial intelligence-driven customised marketing strategies have increased sales and started conversion rates to as much as 30%. The findings show that to meet customer demands and enhance their experience, devotion, and corporate success, AI technologies are essential in CRM systems. Future studies should examine long-term patterns in AI-CRM adoption as well as moral dilemmas about openness and data protection. By incorporating these elements into AI-powered platforms, companies may maximise customer encounters and relationships while maintaining consumer trust.

1. Ahmad, R. and Pande, S., 2024. Examining the Integration of Artificial Intelligence in Customer Relationship Management. Available at SSRN 4816041. https://scholar.googleusercontent.com/scholar?q=cache:caw2bxr4BQ8J:scholar.google.com/+Conversational+AI+in+Customer+Relationship+Management:+Harnessing+Deep+Learning,+Cloud+Computing,+and+Real-Time+Data+Analytics&hl=en&as_sdt=0,5&as_ylo=2021
2. Ahsan, M.M., Mahmud, M.P., Saha, P.K., Gupta, K.D. and Siddique, Z., 2021. Effect of data scaling methods on machine learning algorithms and model performance. Technologies, 9(3), p.52.https://www.mdpi.com/2227-7080/9/3/52
3. Alizadegan, H., Rashidi Malki, B., Radmehr, A., Karimi, H. and Ilani, M.A., 2025. Comparative study of long short-term memory (LSTM), bidirectional LSTM, and traditional machine learning approaches for energy consumption prediction. Energy Exploration & Exploitation, 43(1), pp.281-301.https://journals.sagepub.com/doi/abs/10.1177/01445987241269496
4. Alladi, R., 2024. Harnessing the Power of Gen AI & Cloud Computing for Customer Relationship Management. International Journal of Scientific Research & Engineering Trends, 10(3), pp.724-730. https://scholar.googleusercontent.com/scholar?q=cache:TPlntn450tYJ:scholar.google.com/+Conversational+AI+in+Customer+Relationship+Management:+Harnessing+Deep+Learning,+Cloud+Computing,+and+Real-Time+Data+Analytics&hl=en&as_sdt=0,5&as_ylo=2021
5. Ansari, A. and Tabassum, T., 2024. Customer Relationship Management Using AI Tools in Management System. Journal of Scientific Research and Technology, pp.103-118. https://scholar.googleusercontent.com/scholar?q=cache:1-QsdSMRD4cJ:scholar.google.com/+Conversational+AI+in+Customer+Relationship+Management:+Harnessing+Deep+Learning,+Cloud+Computing,+and+Real-Time+Data+Analytics&hl=en&as_sdt=0,5&as_ylo=2021
6. Chicco, D., Warrens, M.J. and Jurman, G., 2021. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. Peerj computer science, 7, p.e623.https://peerj.com/articles/cs-623/
7. Chung, J., Ding, Y. and Kalra, A., 2023. I really know you: how influencers can increase audience engagement by referencing their close social ties. Journal of Consumer Research, 50(4), pp.683-703.https://academic.oup.com/jcr/article-abstract/50/4/683/7077143
8. Egbuhuzor, N.S., Ajayi, A.J., Akhigbe, E.E., Agbede, O.O., Ewim, C.P.M. and Ajiga, D.I., 2021. Cloud-based CRM systems: Revolutionizing customer engagement in the financial sector with artificial intelligence. International Journal of Science and Research Archive, 3(1), pp.215-234. https://scholar.googleusercontent.com/scholar?q=cache:yhjMwv3uthYJ:scholar.google.com/+Conversational+AI+in+Customer+Relationship+Management:+Harnessing+Deep+Learning,+Cloud+Computing,+and+Real-Time+Data+Analytics&hl=en&as_sdt=0,5&as_ylo=2021
9. Elshewey, A.M., Shams, M.Y., Tarek, Z., Megahed, M., El-Kenawy, E.S.M. and El-dosuky, M.A., 2023. Weight Prediction Using the Hybrid Stacked-LSTM Food Selection Model. Syst. Sci. Eng., 46(1), pp.765-781.https://www.researchgate.net/profile/Ahmed-Elshewey/publication/367295795_Weight_Prediction_Using_the_Hybrid_Stacked-LSTM_Food_Selection_Model/links/63caa571d9fb5967c2ef4b2f/Weight-Prediction-Using-the-Hybrid-Stacked-LSTM-Food-Selection-Model.pdf?origin=journalDetail&_tp=eyJwYWdlIjoiam91cm5hbERldGFpbCJ9
10. Fernando, K.R.M. and Tsokos, C.P., 2021. Dynamically weighted balanced loss: class imbalanced learning and confidence calibration of deep neural networks. IEEE Transactions on Neural Networks and Learning Systems, 33(7), pp.2940-2951.https://ieeexplore.ieee.org/abstract/document/9324926/
11. Fionah, M.J., 2024. Harnessing Artificial Intelligence in Customer Relationship Management: Transforming Customer Interactions and Business Performance.
12. Hanif, M. and Best, T., 2024. Forecasting Netflix Stock Prices: A Case Study Using Regression and Machine Learning Models.https://www.researchgate.net/profile/Thomas-Best-4/publication/385417917_Forecasting_Netflix_Stock_Prices_A_Case_Study_Using_Regression_and_Machine_Learning_Models/links/6723c554db208342dee0f867/Forecasting-Netflix-Stock-Prices-A-Case-Study-Using-Regression-and-Machine-Learning-Models.pdf
13. Hu, Z., Zhao, Y. and Khushi, M., 2021. A survey of forex and stock price prediction using deep learning. Applied System Innovation, 4(1), p.9.
14. Hussain, H., Tamizharasan, P.S. and Rahul, C.S., 2022. Design possibilities and challenges of DNN models: a review on the perspective of end devices. Artificial Intelligence Review, pp.1-59.https://link.springer.com/article/10.1007/s10462-022-10138-z
15. Inavolu, S.M., 2024. Exploring AI-driven customer service: Evolution, architectures, opportunities, challenges and future directions. International Journal of Engineering and Advanced Technology, 13(3), pp.156-163. https://scholar.googleusercontent.com/scholar?q=cache:1aK6DdZ08QEJ:scholar.google.com/+Conversational+AI+in+Customer+Relationship+Management:+Harnessing+Deep+Learning,+Cloud+Computing,+and+Real-Time+Data+Analytics&hl=en&as_sdt=0,5&as_ylo=2021
16. Jamil, K., Anwar, A. and Mustafa, S., 2025. Upgradation of Customer Relationship Management Through the AI-Driven Approach to Personalization, Engagement, and Retention. In Reshaping Organizational Management and Workplace Culture With AI (pp. 263-286). IGI Global Scientific Publishing. https://www.igi-global.com/chapter/upgradation-of-customer-relationship-management-through-the-ai-driven-approach-to-personalization-engagement-and-retention/378613
17. Leewayhertz, 2025. AI in data integration: Types, challenges, key AI techniques and future. https://www.leewayhertz.com/ai-in-data-integration/#The-role-of-AI/ML-in-data-integration
18. Mienye, I.D., Swart, T.G. and Obaido, G., 2024. Recurrent neural networks: A comprehensive review of architectures, variants, and applications. Information, 15(9), p.517.https://www.mdpi.com/2078-2489/15/9/517
19. Muhammad, T., Aftab, A.B., Ibrahim, M., Ahsan, M.M., Muhu, M.M., Khan, S.I. and Alam, M.S., 2023. Transformer-based deep learning model for stock price prediction: A case study on Bangladesh stock market. International Journal of Computational Intelligence and Applications, 22(03), p.2350013.
20. Paramesha, M., Rane, N. and Rane, J., 2024. Big data analytics, artificial intelligence, machine learning, internet of things, and blockchain for enhanced business intelligence. Artificial Intelligence, Machine Learning, Internet of Things, and Blockchain for Enhanced Business Intelligence (June 6, 2024). https://scholar.googleusercontent.com/scholar?q=cache:obOg9Og6VdwJ:scholar.google.com/+Conversational+AI+in+Customer+Relationship+Management:+Harnessing+Deep+Learning,+Cloud+Computing,+and+Real-Time+Data+Analytics&hl=en&as_sdt=0,5&as_ylo=2021
21. Peivandizadeh, A., Hatami, S., Nakhjavani, A., Khoshsima, L., Qazani, M.R.C., Haleem, M. and Alizadehsani, R., 2024. Stock market prediction with transductive long short-term memory and social media sentiment analysis. IEEE Access.https://ieeexplore.ieee.org/abstract/document/10528270/
22. Raj, K., Fredrick, D.P., Kurahattidesai, C. and Hegde, C.S., 2024. Artificial Intelligence Driven Customer Relationship Management: Harnessing the power of technology to improve business efficiency. International Journal of Communication Networks and Information Security, 16(4), pp.58-65. https://www.proquest.com/openview/9400c362addeff3925f105bf98e5bbd7/1?pq-origsite=gscholar&cbl=52057
23. Rane, N., 2023. Enhancing customer loyalty through Artificial Intelligence (AI), Internet of Things (IoT), and Big Data technologies: improving customer satisfaction, engagement, relationship, and experience. Internet of Things (IoT), and Big Data Technologies: Improving Customer Satisfaction, Engagement, Relationship, and Experience (October 13, 2023). https://scholar.googleusercontent.com/scholar?q=cache:K8e2le58YVoJ:scholar.google.com/+Conversational+AI+in+Customer+Relationship+Management:+Harnessing+Deep+Learning,+Cloud+Computing,+and+Real-Time+Data+Analytics&hl=en&as_sdt=0,5&as_ylo=2021
24. Santos, C.F.G.D. and Papa, J.P., 2022. Avoiding overfitting: A survey on regularization methods for convolutional neural networks. ACM Computing Surveys (Csur), 54(10s), pp.1-25.https://dl.acm.org/doi/abs/10.1145/3510413
25. Sarkar, S., Vinay, S., Djeddi, C. and Maiti, J., 2022. Classification and pattern extraction of incidents: A deep learning-based approach. Neural Computing and Applications, 34(17), pp.14253-14274.https://link.springer.com/article/10.1007/s00521-021-06780-3
26. Switrayana, I.N., Hammad, R., Irfan, P., Sujaka, T.T. and Nasri, M.H., 2025. Comparative Analysis of Stock Price Prediction Using Deep Learning with Data Scaling Method. JTIM: Jurnal Teknologi Informasi dan Multimedia, 7(1), pp.78-90.https://www.journal.sekawan-org.id/index.php/jtim/article/view/650
27. Kabade S, Sharma A, Kagalkar A. Cloud-Native AI Solutions for Sustainable Pension Investment Strategies. IJAIDSML [Internet]. 2025 Feb. 12 [cited 2025 Sep. 17];6(1):196-204. Available from: https://ijaidsml.org/index.php/ijaidsml/article/download/220/198
28. Sharma A, Kabade S, Kagalkar A. AI-Driven and Cloud-Enabled System for Automated Reconciliation and Regulatory Compliance in Pension Fund Management. IJERET [Internet]. 2024 May 3 [cited 2025 Sep. 17];5(2):65-73. Available from: https://ijeret.org/index.php/ijeret/article/download/234/222