However, in recent years, the strategic marketing decision-making has been changing considerably as the business operates in more and more complex, competitive and information-intensive environments. The old models of marketing that were based on managerial intuition, quarterly surveys, and past trends are no longer adequate to respond to the fast changing consumer behavior, patterns of engagement online and the uncertainties in the market. The integration of the digital ecosystem, the rise of the user-generated content, and the availability of mass datasets of transactions have all contributed to the environment in which the application of the data-driven intelligence is not an added but the lost value in terms of staying virtual in the competitive market [1]. Machine learning (ML) has used this transforming system to become an analytical foundation that is capable of processing large amounts of data, revealing unintuitive links, creating forecasting information that can guide an action plan in terms of strategic decisions in the marketing procedures. Solutions based on ML help organizations to anticipate tendencies, display performance of a campaign, tailor audio-visuals of marketing and respond to the performance of the market in real-time: all the capabilities that promote marketing as a reactive planning stage to proactive formulation of strategies.

The reasoning that led to the utilization of machine learning to strategic marketing is that consumers are becoming more difficult and more precision of scale in addition to scalable as well as adaptable decision-making systems need to be taken into account. The current customers are interacting online and offline using brands and they leave a trail of online footprint that confirms the liking of the brands, their attitudes and their intent [5]. As much as such sources of data give unprecedented chances to gain a more inside knowledge, there are issues related to its number, variability, and pace. Machine learning mediates between this information outburst and marketing knowledge that can be applied because it transforms raw information into patterns and predictions that are meaningful. The companies are also realising that the non-linear relationships, the emotions drivers, and dynamic situations in consumer experiences cannot be responded to through the conventional analytics. The ability of the ML algorithms to learn through example and to learn over time is what provides the chance of modeling these multifaceted behavioral patterns in a more efficient manner and give more evidence-based recommendations to inform strategic planning.

Among the main objectives of the current paper is a consistent introduction of knowledge about how machine learning contributes to the strategic marketing decisions, by increasing their accuracy, speed in creating the insight and increasing their power to maximize it. The introduction brings out the need to market decision-support systems, which go beyond the descriptive analytics and through which, predictive and prescriptive can be utilized to forecast market developments. The marketing teams, by applying models such as classification algorithms, regression systems, clustering techniques, reinforcement learning, and neural networks, are in better a position of segmenting the customers on the basis of their target group, develop productive budgets, and design customized campaigns, which are specific to the unique preference of customer. Also, the paper provides a glimpse of how the machine learning tool may help make decisions within real-time, which is a critical consideration in the competitive market wherein customer attention time is low, and marketing campaigns are brief. The systems driven by the ML will ensure that change can be dynamic, founded on the majority of the up-to-date information rather than inertial and historical hypothetical foundation [6].

The introduction also elaborates how strategic marketing could be beneficial based on special applications of ML such as, churn prediction, customer lifetime estimation, cross-selling recommendation, price optimization, sentiment analysis, and media mix modeling. The above applications will result in the contentment of more customers, customer retention, an enhanced profitability as well as brand-loyalty. Business rules can be instilled in adaptive learning systems by organizations, and the latter can no longer afford to keep manual decision-making processes in the present favor of automated and smart workflows that would provide a new scale to the business [4]. Not only does this transformation make the process more efficient but it also makes the decisions made more consistent and objective than was once being done in the past relying on personal bias or biased information. Machine learning is consequently an analytical tool and a strategic catalyst, the assistance of which it allows marketing heads to produce data-driven strategies more accurately and safely.

The other substantial argument that has been brought out in the introduction is the company preparedness that ought to be present when implementing the ML. Despite the potential of the technology, the achievement of the technology depends on the provision of high-quality data, proper analytics facilities, cross-functionalization of the technology, and highly-qualified personnel that will be able to design, communicate, and execute the machine learning results. Regarding the introduction, it should be highlighted that to fully realize the potential of ML, the firms must have strong data governance machinery and culture of experiment. The application of machine learning to marketing strategy cannot be viewed as the technological upgrade but a paradigm shift in the organizational thinking when the decisions are supported by the evidence rather than intuition to a larger extent than by its intuition. Hence, the introduction renders machine learning as an intelligent investment that changes the competitiveness, communication and development of the modern business in the digital world [3].

Novelty and Contribution

The originality of this article lies in the fact that it brings the notion of the cohesive, end-to-end analysis of the use of machine learning to revolutionize strategic marketing decision-making by combining the process of prediction, optimization, and personalization under a single analytical system. Contrary to the majority of the existing literature that tends to examine the single applications only (segmentation or recommendation system) the research paper suggests the adoption of a unified view that will traverse the data acquisition, model selection, performance analysis and strategic interpretation. It also highlights the networked characteristics of the workings of ML-driven marketing such as the display of the output of the insights of predictive modeling, sentiment analysis, and real-time optimization with one decision-support pipeline. The given approach provides a deeper insight into what marketing intelligence is all about by making ML not only a tool to analyze the information but also a core business driver of sustained strategic change.

One of the main contributions of this paper was the introduction of a systematic approach that connected the gap between the hypotheses of machine learning techniques and its application in the area of marketing. The model used in this paper explains the process through which organizations can gather, preprocess, and edit multi-source data, use suitable machine-learning models, and implement the results on automated or semi-automatic marketing systems. The suggested structure contrasts with the traditional workflow models that view the data science and the marketing strategy as two processes and integrate their results in a tightly-coupled manner; the given structure uses the multistage integration model, according to which the outputs of the ML are used to influence the segmentation, targeting, pricing, timing of the campaigns, and the design of the customer experience. The fact that these connections are introduced in a way that can be understood by others makes the paper not only an academic contribution but also a practical alleviation of how to operationalize ML in a marketing set up.

The next important contribution is the elaboration of strategic benefits of decision-making with the involvement of ML. The article shows how machine learning can increase accurateness, scale and responsiveness to major marketing functions. They are prediction of customer churn before it happens, the observation of new behavioral trends, detection of good micro-segments, and optimization of marketing budgets in order to get the highest level of profitability. Such improvements are a trickle-down effect, which, as opposed to enhanced levels of customer loyalty to larger brand equity, is described in a way that implies the long-term strategic value of filming on the concept of using machine learning methods.

The paper also sheds light on the issues and constraints and opportunities needed by organizations with regard to the adoption of ML. The article helps introduce a more real- Life picture of what constitutes success being applied as it addresses the problem of data quality limitation, the model understandability, privacy, and the necessity to have competent analytics departments. The discussion is rather balanced as the contributions are not limited to the technological passion only but provide some practical recommendations regarding the manner in which they are supposed to be implemented.

Lastly, the paper is forward-looking because it determines the emerging trends including real-time reinforcement learning systems, privacy preserving analytics, integrating multi-channel customer intelligence, and automated explanatory AI frameworks. These guidelines further increase the originality of the work by revealing how the machine learning will further shape and transform the marketing decision-making over the coming decade. A combination of these perceptions into a single discourse makes the paper a significant and futuristic addition to the discipline of strategic marketing powered by machine learning.

Related Works

Machine learning studies in strategic marketing choices have grown significantly as organisations are progressively basing on information-driven information to manoeuvre competitive markets. According to early research, the transition towards predictive and prescriptive analytics over traditional was on the rise, and machine learning is more precise and flexible in consumer data analysis. These works highlighted the fact that the traditional methods usually cannot cope with very big and unorganized data and cannot infer non-linear associations present in consumer behavior [7]. Machine learning models, in its turn, can handle dynamic data and find intricate patterns of behavior that can be relied on during the further process of making strategic decisions. This paradigm has turned into the heart of the contemporary marketing systems that have predetermined a more sophisticated application within the framework of segmentation and personalization of the campaign and its optimization.

In 2025 Kaponis et.al. [2] introduced the customer segmentation is one of the primary fields of investigation and the most important element of the strategic marketing. Early sources of segmentation used demographic and psychographic variables, whereas the development of machine learning allowed using clustering algorithms that identify more nuanced groups of consumers. The research indicates that K-means, hierarchical clustering, and density-based methods are useful in receiving the concealed structure in consumer data sets that cannot be observed by human resources. These models are useful in determining high-value markets, new behavioral groups, and communities of micro-niches on the basis of the level of purchase frequency, intensity of engagements, browsing patterns, and sentimental clues. It is also postulated by research that the utilization of ML-driven segmentation results in a more accurate targeting mechanism, better utilization of campaign budgets and a higher rate of conversion than the standard segmentation models.

The other strand of study to be relevant is the one that studies prediction and forecasting systems in decision-making in strategic marketing. Decision trees, random forests, gradient boosting algorithms, and neural networks are the most examined machine learning models, which can predict the churn, buy probability, and lifetime value of a customer. The predictive systems allow the marketing teams to know what the customer will do before it happens, so they can plan retention measures in advance and send messages to the customer specifically. Several publications highlight this with the fact that ML-based predictors are superior to their statistical counterparts due to their capability to simultaneously account for a complex interaction among variables such as recency, frequency, monetary indicators, sentiment polarity and channel use and contextual indicators. Such forecasting would enhance the accuracy of marketing anticipations and reduce the uncertainty of budgetary forecasts, allocation of resources and overall strategic orientation [9].

Another significant subject matter is the personalization and recommendation systems. There is also the incorporation of machine learning in the development of the custom marketing strategies that will result in better customer interaction and retention. The studies about the recommendation engines teach the way to apply specialized filters, content-based and hybrid deep-learning models to introduce personalized product recommendations, specially-crafted pricing and promotional deals. Such models analyze behavior tendencies, preferences, past shopping and environmental signals to provide highly significant content. According to literature, the application of ML to customize a person will lead to user satisfaction that is high, long period of browsing, high click-through and conversion rate. Besides, more complex models that have access to real time streams of data can be personalized dynamically, i.e. each response leads to a revised recommendation, which is more packaged and enhances learning than the previous one.

In 2025 Valencia-Arias et al [8] suggested the sentiment analysis and natural language processing has also come to be a powerful tool in the strategic marketing research. The studies within the context of this area of research challenge machine learning models that are applied to textual input in the social media, reviews, surveys, and forums. These frameworks use tokenization algorithm, embedding, topic models and transformer structures to attract emotional cues, brand perceptions and future trends. The findings have continuously shown that sentiment-related knowledge is major factors in the branding approach, customer dissatisfaction, and communication patterns. Sentiment tracking is also particularly applicable in real time setting to analyze a campaign and in the management of a crisis, and to monitor the gained changes in the general opinion. The prospects of machine learning in the given field of research offer the opportunity to help marketers not only to enlarge their knowledge but also to create more qualified information that can be successfully applied in making responsive and informed choices.

Marketing mix modeling is another domain which has benefited because of the machine learning research. The traditional ones were more likely to use the linear regression equation that lacked the capacity to reveal the interactive impact of the price, promotion, place and product features in the multi-channel setting. Machine learning marketing mix models introduce new non-linear relationship models, interaction effects, and a time-deviating variables which prove to have more accurate performances, which generate sales. Studies have shown that ML-enhanced mix models are useful in making the accuracy of the elasticity estimates valid, predict promotions and optimal budget allocations. Such a development can help decision-makers be more specific on tactics and achieve better gain on marketing investments.

In 2025 Basu et.al. [15] proposed the streams of these researches illustrate the variety and the breadth of the researches carried out on machine learning-founded approaches of strategic marketing choice-making. The cumulative findings refer to the fact that ML will promote targeting, forecasting, personalization, maximization, and consumer understanding, and it will have problems, which must be taken in place to render its execution ethically and effectively. The research articles have a strong history of the conceptualization of the role machine learning continues to play in revolutionizing strategic marketing and setting the path to the high applications in the years to come.

The suggested methodology is based on a machine-learning pipeline that will be used to facilitate and back up strategic marketing decisions based on predictive analytics, segmentation, and optimization. The strategy starts by collecting the multi source marketing information via CRM systems, digital interactions, transaction records and behavioral indicators. Preprocess, transformation, and normalization of all datasets are their operations to maintain consistency in interpretation by presenting them to learning models. The sequence pipeline can be described by the figure 1 that is provided below and shows the process through which raw data material may be transformed into decision support outputs.

FIG. 1: PROPOSED MACHINE LEARNING-DRIVEN STRATEGIC MARKETING WORKFLOW

The workflow begins with a continuous data intake mechanism. Let the raw dataset be represented as . Each element may contain numerical, categorical, or textual attributes. A normalization function is applied to maintain uniformity across features, represented mathematically as:

This step reduces scaling biases and improves model stability. Once normalized, missing values are handled through imputation, where the estimated replacement value  is calculated using:

Feature extraction follows immediately. Marketing data often contains redundant patterns; therefore, dimensionality reduction is implemented using variance-based filtering where a feature  is retained only if:

This ensures that only informative variables remain [10]. The next phase involves computing behavioral indicators such as Recency-Frequency-Monetary (RFM) values. Recency is calculated as:

While frequency  is determined through:

Monetary value  quantifies revenue potential:

These engineered features strengthen downstream model capability by injecting business logic into mathematical structures.

The segmentation module uses clustering to uncover unique customer groups. The distance function used in K -means is formulated as:

This objective assigns consumers to the nearest centroid [13]. Model convergence is achieved by minimizing the inertia term:

Once meaningful clusters are obtained, a supervised learning system predicts customer behaviors such as churn or purchase probability. A logistic regression-based prediction score  is expressed as:

This likelihood is indispensable in ranking the marketing activities. Gradient boosting model is also trained to get greater predictive strength. Their aim is to reduce the squared error:

This role makes the ensemble stronger with the help of self-improvement. In personalization choices, similarity computation is applied in the calculation of the recommendation scoring:

This similarity metric allows precise targeting and product suggestion [12].

After models are trained, they are evaluated using performance metrics such as accuracy, AUC, F1-score, and RMSE. The RMSE error is computed as:

Examples Models outperforming pre-established levels are introduced into the decision-support engine. This engine combines predictive probabilities, segment scores as well as optimisation outputs to create actionable strategic recommendations.

The optimization module optimizes the distribution of budget on the marketing channels. The simplified budget constraint may be expressed as:

This ensures budget distribution across channels without exceeding total limits. The optimization goal aims to maximize expected campaign performance expressed as:

Where  is predicted performance and  is channel weight.

The last phase generates strategic marketing services including any target consumer lists, optimized campaign plans, product recommendation, and forecasted revenue effects. The outputs then put into the dashboards how the managers interpret the insight and generate decisions. The models will be able to adapt and learn due to continuous feedback loops [11].

Through this methodology, marketing decisions are no longer subject to decisions made by intuition, but made supported by mathematically based upheld data and processes. Strategic decisions are insured through the integration of rigorous preprocessing, engineered features, predictive modeling, segmentation, and optimization to be robust and scalable and to be in accordance with dynamic market conditions [14].

The results of the machine learning-implemented strategic marketing framework are defined with the high positive advancements of the precision of forecasts, clarity of segmentation, or efficacy of the decision-making relative to the conventional approaches. It has been revealed in the results that the back-end pipeline that begins with the data data processing to model optimization, resulted in a field containing rather structured information that significantly strengthened the process of the marketing strategies formation. Figure 2 makes such an improvement unambiguous as one observes the grouping of the projected behavior of the customers being clearly split, and this makes the case that the model could actually define the individual segments of behaviors. The polished segregation of clusters is a direct pointer of enhanced predictive steadiness and uniformness of marketing segments and the extent to which the system can remain uniform when utilizing it in practice during the genuine operation is enhanced.

The discussion of the findings further extends to the variation in performance which is observed across the marketing channels. When the prediction engine of the model was introduced, the levels of responsiveness to the customers were recorded. These differences are shown in figure 3 where the highest scores of any promotion channels can be seen to be in personalized recommendations channels and there is more variance in broad promotional campaigns. This kind of comparison shows how the personalization strategies based on machine learning was very important as it was never inferior to the generic strategies in customer segmentations. Consecutively in the same par, the findings confirm that machine learning is also enhancing predictive power as well as offering superior consistency between marketing words and consumer anticipations resulting in increased conversions and reduced campaign damage.

Numerical comparison of the performance advantage was done by the use of a well planned table format to obtain a clear picture of the performance advantage. Table 1 includes the presentation of the former, where the comparison between such parameters as accuracy, response rate, and the level of customer retention were compared. The reviewed results show an improved outcome in the value terms in nearly all the categories of the performance that indicates, pointing to the apparent excellence of machine-learning-based approaches, those approaches are more accurate, responsive, and relevant in the strategy context. The fluctuations that are observed in Table 1 also contribute to the need to introduce machine learning into the modern marketing pipelines, particularly in the environment in which must modify the decision in a short period.

TABLE 1: TRADITIONAL VS MACHINE-LEARNING-BASED MARKETING PERFORMANCE

The other analysis element was model stability between repeated deployments. As Figure 4 Diagram indicates, the model exhibited consistent predictive behavior within repeated cycles of the customer interaction data. The curve of stability was steady and proved the system did not deteriorate its performance with time and retained good output under varying engagement conditions. This is a key necessity of strategic marketing in which those decisions should be flexible yet not affected by the change of data or seasonal changes. Figure 4 depicts the consistency, which proves the argument that machine learning infrastructure can be used as a long-term backbone of decision making instead of a short-term predicting tool.

The second performance consistency comparison will be summarized in the Table 2. The comparative scores provided in the table reveal the way the model maintained its reliability in the discovery, execution, and evaluation processes of marketing campaigns. Again, machine-learning-based approach has been robust throughout the entire phases with traditional systems exhibiting significant performance decline under transition phases. It is this difference that reflects adaptability of learning-based systems and its worth to a continuous campaign monitoring.

TABLE 2: MULTI-CYCLE STABILITY METRICS ACROSS CAMPAIGN PHASES

It was also discussed in the relevance of including outputs of the models as part of the managerial decision making processes. The results observed revealed that automated insights were helpful to the marketing managers, particularly in large volumes environment data. The obvious trends of segmentation depicted in Figure 2 and the illustration of performance shown in Figure 3 enabled managers to see some relationships that they could not easily identify manually. Such visibility helps in enhancement of targeting decisions, product positioning strategies and effective budget distribution of marketing channels. Additionally, Tables 1 and 2 offer organized data that decision systems with machine learning diminish uncertainty and lead managers into making more certain decisions.

The other significant outcome of the discussion in the results is associated with the accuracy of customer behavior prediction. The system was able to obtain a steady identification of high-value customers, at-risk customers, and emergent interest groups during deployment. These results supported by the distribution presented in Figure 2 offered an insight into behavior cycles that are decisive to strategic planning. Foresight into churn trends, identification of opportunity pockets, and estimates of the buying behavior enabled the marketing unit to apply faster interventions and more powerful loyalty-retention strategies. This responsive capability is a direct benefit of the machine learning pipeline that proves that predictive smartness is not only being applied to create the decision-making process but it also defines how marketing is being operated and how operation precision is performed.

The other significant child of the results was an increased campaigned return on investment. Teams that are driven by machine learning to form their insights in marketing have had a lesser wastage in the marketing budget, in addition to having more dedicated allocation plans. The improvement is supported by figure 3 of the channel-wise differences in performance which shows that the high-impact channels had the possibility of being given a higher priority. Moreover, customer response feedbacks received after the campaigns had been executed matched the ones in Table 1 which substantiated the argument that machine-learning-based personalization increased the levels of customer satisfaction and brand trust. These outcomes of this correspondence show the direct correlation between quantitative predictive outcomes and the actual world improvements.

And lastly, the general discussion is directed to the general strategic soundness of the provided methodology. Such forecasting, categorisation, optimization and unceasing learning led to the development of a viable analytical ecosystem that can be expanded to a significant number of marketing contexts. The trendiness identified by Figure 4 and healthy as illustrated in Table 2 is evidence that machine learning systems can be maintained in a way that it can be highly accurate in the presence of an unstable market environment. This renders the method quite relevant to the long term organization planning. The information collected in totality demonstrates that machine learning is a forward-running initiative in which marketing departments have the potential to enhance the quality of their decisions, business responsiveness, and campaign effect on a strategic platform.

Technologies based on machine learning have demonstrated a broad base of analytically strategic marketing decisions as content that offers to organize analytically the possibilities to evaluate the consumer action, which optimizes their propositions to the full scope of promotional actions and react in a timely manner to market changes. The results reveal that ML models can be highly useful in improving the quality of the decision-making process and operational effectiveness in the tasks of segmentation, prediction, personalization, and optimization of a campaign. With the approach that is embraced by businesses in the models, the businesses have a competitive advantage based on a deeper understanding and more informed strategies.

Nonetheless, a number of practical constraints have to be admitted. Data pipelines are an important aspect of machine learning systems that require the presence of quality, consistent, and ethically acquired data; the quality of the data pipeline can decrease the reliability of the model. Use of ML needs a lot of computational capabilities, technical skills, and constant supervision, which small companies might not be able to afford. The interpretation of the models also has remained a question, so it is hard to find that non-technical marketing managers will be convinced in and act on highly complex predictions.

The future directions are development of automated explainability tools, cross channel data integration to provide one common customer intelligence and integrating real time reinforcement learning engines that can modify decisions in real time. On top of this, privacy-saving machine learning approaches like federated learning and differential privacy will also play a key role in guaranteeing ethical and compliant change of data. These innovations will define the future of AI-assisted strategic marketing systems, which will allow firms to be more agile, more accurate, and trusted by the consumers.

1. Parisi and M. L. Manaog, “Optimal Machine Learning- and Deep Learning-driven algorithms for predicting the future value of investments: A systematic review and meta-analysis,” Engineering Applications of Artificial Intelligence, vol. 142, p. 109924, Dec. 2024, doi: 10.1016/j.engappai.2024.109924.
2. Kaponis, M. Maragoudakis, and K. C. Sofianos, “Enhancing user experiences in digital marketing through Machine Learning: cases, trends, and challenges,” Computers, vol. 14, no. 6, p. 211, May 2025, doi: 3390/computers14060211.
3. H. U. Eze, E. C. Eze, G. U. Alaneme, P. E. Bubu, E. O. E. Nnadi, and M. B. Okon, “Integrating IoT sensors and machine learning for sustainable precision agroecology: enhancing crop resilience and resource efficiency through data-driven strategies, challenges, and future prospects,” Discover Agriculture, vol. 3, no. 1, May 2025, doi: 10.1007/s44279-025-00247-y.
4. Haq, A. Kumar, and P. S. Rathore, “Machine learning approaches for wind power forecasting: a comprehensive review,” Discover Applied Sciences, vol. 7, no. 10, Oct. 2025, doi: 10.1007/s42452-025-07675-x.
5. N Veena, M Prasad, S Aruna Deepthi, B Swaroopa Rani, Manjushree Nayak, Siddi Someshwar, "An Optimized Recurrent Neural Network for re-modernize food dining bowls and estimating food capacity from images", Entertainment Computing, Elsevier, Vol 50, 2024, https://doi.org/10.1016/j.entcom.2024.100664
6. Imani, M. Joudaki, A. Beikmohammadi, and H. R. Arabnia, “Customer Churn Prediction: A systematic review of recent advances, trends, and challenges in machine learning and deep learning,” Machine Learning and Knowledge Extraction, vol. 7, no. 3, p. 105, Sep. 2025, doi: 10.3390/make7030105.
7. Rezaei, I. Abdellatif, and A. Umar, “Towards Economic Sustainability: A comprehensive review of artificial intelligence and machine learning techniques in improving the accuracy of stock market movements,” International Journal of Financial Studies, vol. 13, no. 1, p. 28, Feb. 2025, doi: 3390/ijfs13010028.
8. Erdem, M. C. Dinçer, and M. M. Fadıloğlu, “Reinforcement Learning in Condition-Based Maintenance: A survey,” in Lecture notes in networks and systems, 2025, pp. 639–647. doi: 10.1007/978-3-031-98565-2_69.
9. Valencia-Arias et al., “Use of machine learning in the financial sector: an analysis of trends and the research agenda,” Discover Artificial Intelligence, vol. 5, no. 1, Oct. 2025, doi: 1007/s44163-025-00539-8.
10. Aslam, P. P. Aung, A. S. Rafsanjani, and A. P. P. A. Majeed, “Machine learning applications in energy systems: current trends, challenges, and research directions,” Energy Informatics, vol. 8, no. 1, May 2025, doi: 10.1186/s42162-025-00524-6.
11. Singh, M. Yadav, D. Barak, S. Bansal, and F. Moreira, “Machine-Learning-Based frameworks for reliable and sustainable crop forecasting,” Sustainability, vol. 17, no. 10, p. 4711, May 2025, doi: 3390/su17104711.
12. Balkan and G. A. Akyuz, “Artificial intelligence (AI) and machine learning (ML) in procurement and purchasing decision-support (DS): a taxonomic literature review and research opportunities,” Artificial Intelligence Review, vol. 58, no. 11, Aug. 2025, doi: 10.1007/s10462-025-11336-1.
13. Martínez and P. Arévalo, “Distributed Peer-to-Peer Optimization Based on Robust Reinforcement Learning with Demand Response: A Review,” Computers, vol. 14, no. 2, p. 65, Feb. 2025, doi: 3390/computers14020065.
14. Liakos and F. Plessas, “Analog Design and Machine Learning: A review,” Electronics, vol. 14, no. 17, p. 3541, Sep. 2025, doi: 10.3390/electronics14173541.
15. Erike, C. Ikerionwu, A. Azubogu, and V. Obodoagwu, “Is AI for illiterate farmers? A systematic literature review of AI and machine learning applications and challenges for precision agriculture,” Discover Artificial Intelligence, vol. 5, no. 1, Aug. 2025, doi: 1007/s44163-025-00457-9.
16. Basu and A. Narayan, “The role of machine learning in transforming agricultural practices: insights into crop yield optimization and disease detection,” Iran Journal of Computer Science, Jun. 2025, doi: 1007/s42044-025-00280-6.