Alfred Thaga Kgopa

Water management is a critical challenge in agriculture, particularly for small-scale farms that face resource limitations and unpredictable environmental conditions. Smart irrigation technologies that integrate the Internet of Things (IoT) and machine learning offer significant solutions in enhancing water efficiency and boosting crop production. This study investigates the synergistic application of IoT-enabled sensors alongside machine learning methodologies, specifically Decision Trees (DT) and Support Vector Machines (SVM), to augment irrigation effectiveness. Real-time sensor data collection, featuring elements like soil moisture, temperature, and humidity, serves to direct irrigation techniques. The proposed utilizes solution supervised learning techniques to establish optimal irrigation timetable and reinforcement learning to modify decisions based on real-world performance. Preliminary findings suggest that SVM outperforms DT in reducing false positives and negatives, leading to more precise irrigation control. The study underlines the benefits of AI-driven irrigation system, such as enhanced water conservation, higher crop yields, and increased sustainability. Furthermore, the difficulties of establishing IoT-based irrigation systems, such as data security, connectivity constraints, and cost considerations, are addressed. The findings add to the literature of precision agriculture and provide useful insights for small-scale farmers who are willing to implement smart irrigation solutions. The study's goal is to enhance efficient water use, strengthen food security, and support sustainable farming methods by combining IoT and AI. To get the most out of AI-powered irrigation systems, future research should focus on enhancing algorithm accuracy, expanding real-world trials, and tackling scalability challenges.

Artificial Intelligence (AI)-driven chatbots and intelligent agents are increasingly deployed to support monitoring, evaluation, and organisational learning. Advances in large language models, retrieval-augmented generation, and multi-agent architectures have expanded conversational AI. Despite growing adoption, existing research remains fragmented across technical, educational, and organisational domains, limiting holistic understanding of their design, impact, and governance. This gap creates challenges for organisations seeking evidence-based guidance for implementation. The purpose of this study is to examine and synthesise existing literature on the design, adoption, and impact of AI-based chatbots and intelligent agents within the contexts of monitoring, evaluation, and internal organisational operations. This study presents a systematic literature review and bibliometric analysis of peer-reviewed studies (2021-2025). The review analyses publication trends, core techniques, and application areas of AI-driven chatbots and intelligent agents. Findings reveal rapid growth and increasing focus on organisational learning and evaluation use cases. The study contributes a consolidated synthesis of techniques, benefits, and challenges, identifies research gaps, and offers directions for future research and evidence-based adoption of conversational AI in organisational environments.

– This study developed a neural network model for predicting cattle prices in South African livestock auctions based on breed (B), weight (W), time/season (T) and price (P) variables. Using the online auction dataset from 2023 - 2025, the model analyzed nonlinear relationships influencing price fluctuations, producing realistic per-cattle predictions ranging between ZAR 7,000 - ZAR 17,500, with projected increases up to ZAR 22,000 in future weeks. The results demonstrate the model’s capacity to capture market dynamics shaped by breed attributes, seasonal demand fluctuations, and animal mass. The results illustrate that artificial intelligence-led techniques, including neural networks, can substantially improve market prediction accuracy, enhance profitability, and inform strategic decisions in the livestock sector. Furthermore, this study provides a foundation for future research to expand predictive modelling beyond cattle, contributing to the development of a comprehensive livestock price prediction system that integrates multiple animal types under a unified intelligent forecasting framework.

Plant diseases pose a significant threat to global agricultural productivity, making early and accurate detection critical for yield protection and food security. This study evaluates the evolution, effectiveness, and practical applicability of Machine Learning (ML) and Deep Learning (DL) models for plant disease detection while analyzing research trends to identify leading models, data limitations, and implementation challenges. A systematic literature review and bibliometric analysis were conducted using the PRISMA framework, examining 625 peer-reviewed articles published between 2017 and 2025 from major databases. The analysis highlights the most influential studies, commonly used datasets, and top-performing ML/DL models, assessed in terms of accuracy, methodology, dataset type, and real-time deployment potential. Results show that models such as YOLOv4, VGG19, ResNet50, and MobileNetV2 achieved accuracy levels between 98% and 99.99%, with most trained on the PlantVillage dataset or custom annotated datasets. Several studies demonstrated successful real-time deployment via mobile and edge-device applications. However, key challenges remain, including limited dataset diversity, poor model generalization across environments, and reduced performance under real-field conditions. This study provides a comprehensive overview of progress in AI-based plant disease detection, emphasizing the need for lightweight, adaptable, and field-ready models to support scalable real-world deployment.