International Journal of Professional Practice

Deep Learning Approach for Detection and Prediction of Pest Infections on Plants in Greenhouses

Bridgite Sambu — 2025-11-20

Pest infestations remain problematic in greenhouse agriculture, lowering yields and increasing costs. Manual pest monitoring is laborious, slow, and error-prone, resulting in delayed interventions and excessive pesticide applications. The main objective of this research was to develop an AI-driven hybrid deep learning model for automated pest detection and outbreak prediction, integrating Convolutional Neural Networks (CNNs) for image-based classification and Long Short-Term Memory (LSTM) networks for forecasting to improve response efficiency. This study relied on secondary datasets, such as PlantVillage and IP02, owing to constraints in obtaining primary datasets. These datasets provided pre-annotated pest images and historical climate trends, guaranteeing model robustness. Although the model was trained using secondary data, the study was contextualized by greenhouses in Limuru, Naivasha, and Thika; areas where pest control is an ongoing challenge. To enhance the model's ability to generalize and perform well in an array of agricultural environments, a stratified sampling method which considered farm size and agroclimatic differences was used. Techniques such as image augmentation, noise reduction, and normalization of features were utilized to further improve the quality of the data before the model was trained. Model training and optimization were performed in a GPU-enabled Google Colab environment, which supported batch processing, early stopping, and fine-tuning of hyperparameters. The hybrid model achieved 94.7% accuracy, 93.6% precision, 92.8% recall, and a 93.2% F1-score. With a Mean Absolute Error (MAE) of 0.14 and an R² score of 0.89, the LSTM forecasting module demonstrated its efficiency. This hybrid approach enables early pest identification, preventative actions, and reduced pesticide use.

An Ensembled Tabnet-Based Model Approach for Diabetes Disease Classification

Duncan Ogindo Obunge — 2025-11-20

Despite the advancements in machine learning (ML) for classification tasks, accurately classifying diseases on limited-feature medical datasets remains challenging. Traditional ML models struggle with interpretability, necessitating an exploration of novel technique. This research developed and evaluated a novel TabNet-based ensemble model for diabetes classification, rating its performance against Extreme Gradient Boosting (XGBoost), Random Forest and base TabNet models. The study utilized the PIMA Indian Diabetes dataset from a public ML Repository, which contains 768 tuples (8 features and 1 outcome variable). A TabNet-based ensemble model was developed using a weighted averaging strategy. For comparative analysis, baseline models, including XGBoost, Random Forest, and a standalone TabNet model were also implemented and optimized. Model performance was assessed using key metrics: balanced accuracy, precision and recall (class 1), F1 score, and Receiver Operating Characteristic-Area Under the Curve (ROC-AUC). The ensembled TabNet-based model consistently achieved the highest performance metrics: balanced accuracy of 83%, precision of 84%, recall of 89%, F1 score of 84%, and ROC-AUC of 90.4% compared to XGBoost (accuracy 81% , precision 79% , recall 86%, F1 score 81%, ROC-AUC 88.6%) , Random Forest (accuracy 81%, precision 78%, recall 87%, F1 score 81%, ROC-AUC 91.6%) and base TabNet (accuracy 81%, precision 80%, recall 82%, F1 score 81%, ROC-AUC 86.7%). The study recommends healthcare institutions to adopt the validated ensemble TabNet-based architecture as a standardized framework for clinical decision support systems across multiple diseases. Further, researchers should establish this methodology as the preferred approach for limited-feature medical datasets, extending beyond diabetes to include cardiovascular, hypertension, and cancer screening applications.

The Relationship Between Mentorship and Performance of Ministry of Interior and National Administration Employees in Kajiado County

Mary Wangui Kuria — 2025-11-20

The Government of Kenya has continued to implement public service reforms to enhance the efficiency and effectiveness of service delivery. A critical aspect of this reform agenda is motivating public servants through sustainable, non-cash incentives, in line with cost-cutting measures. This study explored how mentorship, as a form of non-monetary incentive, affects the performance of Ministry of Interior and National Administration employees in Kajiado County. Guided by Herzberg’s Two-Factor, Expectancy, Equity, and Social Learning Theories, the study targeted National Government Administration Officers and selected a sample of 222 respondents through stratified sampling. Data were collected through self-administered questionnaires and analyzed using regression techniques in SPSS (version 27). The findings revealed that mentorship significantly enhances employee performance by providing support, skill transfer, and constructive feedback. Structured mentorship programs were shown to improve competencies, adaptability, and motivation, thereby strengthening service delivery. The study concludes that mentorship is a critical non-monetary incentive that can foster productivity and commitment within the Ministry of Interior and National Administration. It recommends that the Ministry institutionalize structured mentorship programs, including one-on-one coaching, peer learning, and professional guidance, to improve employee capacity and sustain performance improvements.

The Influence of leadership support on project performance in health facilities funded by county governments in the North Rift, Kenya

Mark Amiyo Eyanae — 2025-11-20

Health facility projects in the North Rift, Kenya, often suffer from delays, poor implementation, and inefficient resource use despite the devolved funding system. The purpose of the study was to assess the influence of leadership support on project performance in health facilities funded by the county government in the North Rift, Kenya. The specific objective was to investigate the influence of leadership support on project performance in health facilities funded by the county government in the North Rift, Kenya. The study, guided by transformational leadership theories, used a mixed-methods design. Slovin’s formula was used to obtain a sample of 164 respondents from a population of 282. Data were analyzed using descriptive statistics, correlation, and regression, with results presented in tables and graphs for clarity. The study revealed that leadership support showed a strong correlation (r = .830; p = .000) and a moderate but significant effect (β = .167; p = .049). The study recommended enhancing leadership capacity to improve project performance. Further research was suggested to determine the influence of leadership support on project performance in other public sectors, such as education, water, or infrastructure, to identify cross-sectoral lessons.

Determinants of COVID-19 Vaccine Uptake among Boda-Boda Riders in Mathare, Kenya

John Gershom Otieno — 2025-11-20

Coronavirus disease (COVID-2019) is a novel, highly infectious disease rapidly spreading worldwide. Kenya began the immunization campaign at the beginning of March 2021 with AstraZeneca vaccines, and by June 2022, a total of 18.5 million doses of the coronavirus vaccine had been administered. Due to low uptake of the Coronavirus Disease 2019 vaccine, the Kenyan government called for prioritizing interventions to improve vaccine uptake. This study examined the influence of perception, prior exposure to coronavirus, policy, individual factors, and access to Coronavirus Disease 2019 vaccines on uptake of COVID-19 vaccination among Boda-Boda cyclists in Mathare Sub-County, Nairobi. This was a cross-sectional study with a sample size of 138 Boda-Boda cyclists. The data were purely quantitative and were collected using a structured questionnaire. Simple random sampling was used to select respondents. Data were analyzed using SPSS version 21, and both descriptive and inferential statistics (chi-square, logistic regression) were used. The study found that only a quarter (35 [25%]) of respondents had been vaccinated, of whom 84 (60%) had received the first dose and 56 (40%) had received the required 2 doses. A third (47 [33.6%]) of vaccinated respondents preferred the Johnson & Johnson vaccine. Regarding how policies affected uptake of the COVID-19 vaccine, 97 (69.3%) disagreed that they got vaccinated because they wanted to receive services from government offices. Regarding vaccine accessibility, 78 (55.7%) of respondents agreed that they could get vaccinated at their local health facilities and at any time. The coefficients associated with individual factors, perceptions, policies, and vaccine accessibility were statistically significant (P < .05). The study recommends conducting more educational campaigns on the illness and the need for vaccination.

Factors Associated with Diet Quality of Mothers and Birth Weight of Infants at Lodwar Referral Hospital in Turkana County, Kenya

John Erot Ekiru — 2025-11-20

Maternal factors have been shown to affect maternal outcomes over the years. Low birth weight and preterm birth have been linked to maternal dietary intake during gestation. The prevalence of low birth weight at Lodwar Referral Hospital is 14%. The main objective was to determine the factors associated with maternal diet quality and infants' birth weight in Turkana County. The study used a longitudinal design to collect data from 2023 to 2024 at Lodwar Referral Hospital among mothers aged 18 years and above. The researcher recruited 540 mothers, of whom 500 completed enrollments using a systematic random technique. After pregnancies, 38 defaulted from follow-up, and 2 had stillbirths, resulting in an overall response rate of 93% among mothers. Data were collected using pretested structured questionnaires. SPSS version 29 was used for analysis. A log-binomial model was used to estimate the adjusted risk ratio and its 95% CI for the risk factors for low birth weight. Multi-collinearity was assessed using the variance inflation factor (VIF) with a cut-off of 8; no multicollinearity was found. The overall incidence of low birth weight was 14% (95% CI: 11.1, 17.4%). The difference in low-birth-weight incidence was statistically significant (p-value = 0.006). The risk factors for low birth weight were maternal illiteracy (ARR: 1.8, 95% CI: 1.01, 3.3) and low monthly family income <5000 Ksh. (ARR:1.6,95%CI:1.07,2.2), food taboos during pregnancy (ARR:0.47,95%CI:0.28,0.78), and diet meal number<5 (ARR:1.9,95%CI:1.05,2.61). The prevalence of low birth weight was 14% (70). Low birth weight significantly affected children of mothers with poor diet quality, illiteracy, and poverty. The study recommended that MOH nutritionists promote knowledge of recommended diet quality and exclusive breastfeeding among pregnant mothers. Further research should be conducted to determine the cause of the high prevalence of low-birth-weight cases at Lodwar hospital.

The Extent of Information Communication and Technology Use in Service Delivery among Civil Servants in Meru County

Dorothy Wanja Runji — 2025-11-20

Information and communication technology (ICT) plays an important role in enhancing service delivery. For this reason, the Meru County Government has substantial ICT infrastructure in place, including telecommunications networks, computer systems, internet connectivity, and mobile phone service. However, county government customers cite dissatisfaction with the quality of public service delivery. The study aimed to assess the extent to which information and communication technology (ICT) is utilized to enhance service delivery among civil servants in Meru County, Kenya. It adopted the SERVQUAL model to underpin service delivery. The research also adopted a descriptive design. The target population included 307 employees, encompassing both top-level and middle-level managers across various county departments. The recommendation of 10-30% by Mugenda and Mugenda was used, and 30% adoption led to a sample size of 92. Stratified and simple random sampling techniques were used to obtain 92 employees, of whom 67 were staff members and 25 were administrators. Data collection was carried out using questionnaires. Piloting of the research instruments was done to ensure their reliability. The collected data were analyzed using SPSS version 23, and the results were presented descriptively, including tables and figures. The results indicated that the use of computers and communication media affects overall service delivery in county government departments. However, some members' infrequent use of ICT tools, gaps in data management practices, unreliable internet connectivity, outdated ICT equipment, inadequate ICT proficiency, and security practices were reported. The paper recommends that the Meru County Government, in conjunction with the ICT department, broaden integration, modernize the ICT infrastructure, address connectivity problems, create ICT professional development training programs, and strengthen security measures to optimize its use.