In recent years, Artificial Intelligence (AI) has emerged as a disruptive force, transforming industries worldwide through its ability to analyze vast amounts of data and generate informed decisions. The agricultural sector is undergoing a revolutionary transformation driven by the adoption of this technology to enhance productivity, efficiency, and sustainability. However, the lack of transparency in AI models has sparked doubts about their dependability, particularly when applied to critical industries like medicine and agriculture. To address these challenges, research into Explainable Artificial Intelligence (XAI) gained momentum in early 2017. Tools like Local Interpretable Model-agnostic Explanations (LIME), SHapley Additive Explanations (SHAP), and Gradient-weighted Class Activation Mapping (Grad-CAM) helped make progress. These methodologies have proven instrumental in elucidating the internal workings of complex AI models, making their decision-making processes more transparent and comprehensible. This review surveyed current agricultural developments, including using AI to identify plant diseases. We explored articles from Scopus, Web of Science, IEEE Xplore, PubMed, Google Scholar, Agricola, and the Association for Computing Machinery, published between 2017 and 2024. After extracting and filtering, we included publications that satisfied the selection criteria in the review. It has provided an overview of the uses of XAI methods and the difficulties researchers have encountered. In addition, we reviewed recent advancements in machine learning (ML) and deep learning (DL) to find the limitations that agriculturists face in agriculture. We also provide a modern explainable model for identifying plant diseases and show how XAI can be used in agricultural applications. The conclusion of this paper indicates how XAI is necessary for ML, DL models as well as real time monitoring in the agriculture field.

In this work, we examine the privacy and safety issues of Internet of Things (IoT)-based Precision Agriculture (PA), which could lead to the problem that industry is currently experiencing as a result of Intellectual Property Theft (IPT). Increasing IoT-based information flow in PA will make a system less secure if a proper security mechanism is not ensured. Shortly, IoT will transform everyday lives with its applications. Intellectual Property (IP) is another important concept of an intelligent farming system. If the IP of a wise farming system leaks, it damages all intellectual ideas like cultivation patterns, plant variety rights, and IoT generated information of IoT-based PA. Thus, we proposed an IoT enabled SDN gateway regulatory system that ensures control of a foreign device without having access to sensitive farm information. Most of the farm uses its devices without the use of its integrated management and memory unit. An SDN-based structure to solve IP theft in precision farming has been proposed. In our proposed concept, a control system integrates with the cloud server, which is called the control hub. This hub will carry out the overall PA monitoring system. By hiring the farm devices in the agricultural system, these devices must be tailored according to our systems. Therefore, our proposed PA is a management system for all controllable inputs. The overall goal is to increase the probability of profit and reduce the likelihood of IPT. It does not only give more information but also improves information securely by enhancing the overall performance of PA. Our proposed PA architecture has been measured based on the throughput, round trip time, jitter, packet error rate, and the cumulative distribution function. Our achieved results reduced around (1.66–6.46)% compared to the previous research. In the future, blockchain will be integrated with this proposed architecture for further implementation.

Text vectorization, features extraction and machine learning algorithms play a vital role to the field of sentiment classification. Accuracy of sentiment classification varies depending on various machine learning approaches, vectorization models and features extraction methods. This paper represents multiple ways of evaluations with the necessary steps needed to achieve highest accuracy for classifying the sentiment of reviews. We apply two n-gram vectorization models - Unigram and Bigram individually. Later on, we also apply features extraction method TF-IDF with Unigram and Bigram respectively. Five ensemble machine learning algorithms namely Random Forest (RF), Extra Tree (ET), Bagging Classifier (BC), Ada Boost (ADA) and Gradient Boost (GB) are used here. The key findings in this study is to determine which combination of vectorization models (Bigram, Unigram) along with feature extraction method (TF-IDF) and ensemble classifier gives the better performance of sentiment classification.

Customer satisfaction (CSAT) is vital in service and marketing, indicating how well products or services meet customer expectations. Traditional CSAT methods like the American Customer Satisfaction Index (ACSI) and Net Promoter Score (NPS) face challenges such as survey fatigue and low response rates. This study introduces a novel framework using advanced machine learning (ML) and deep learning (DL) techniques, specifically Bidirectional Encoder Representations from Transformers (BERT), to classify customer feedback into distinct CSAT drivers. Integrating term frequency-inverse document frequency (TF-IDF) methods with BERT-based embeddings, the framework significantly improves prediction accuracy. Using a proprietary dataset of 5,943 customer feedback responses from 39 companies across 13 industries, the fine-tuned BERT model achieved an F1 score of 0.84, surpassing traditional methods like TF-IDF and support vector machine (SVM) with an F1 score of 0.47, and TF-IDF with multi-layer perceptron (MLP) networks at 0.50. A hybrid approach combining BERT and TF-IDF embeddings with MLP networks yielded an F1 score of 0.71. The results show the transformative potential of DL techniques, particularly fine-tuned BERT models, in enhancing CSAT prediction accuracy. This research bridges the gap between traditional and advanced text mining methods, setting a new standard for CSAT modeling and offering a robust framework for extracting actionable insights from customer feedback. It highlights the importance of adopting advanced ML and DL models for strategic decision-making and improving customer satisfaction measurement.