This study aims to evaluate the health resource distribution of provinces in Turkey using DBSCAN cluster analysis method. The optimum values of DBSCAN parameters (epsilon and minPts) were tested by simulation and the clustering silhouette value was taken as the basis for selecting the appropriate parameter set. The results of the descriptive statistical analysis of the dataset show a high coefficient of variation, indicating inequalities in the distribution of health resources. By dividing provinces into two clusters, the study reveals the similarity of local dynamics in the inequality of resource distribution. The findings provide important insights for relevant stakeholders to address the disparities between provinces in Turkey. The fact that the study adopts a method other than the hierarchical and k-means clustering methods dominant in the literature and that the codes of the algorithm are shared in Python language broadens the horizons of the relevant researchers and increases the transparency and reproducibility of the study.
Managing perishable goods, such as food and medicine, is challenging due to their short shelf life and the need for high quality. Traditional methods often fail, leading to overstocking, stock-outs, and spoilage. This study explores how artificial intelligence (AI) can transform this field using machine learning, predictive analytics, and computer vision. AI improves demand forecasting by analysing data such as past sales and weather patterns. Predictive analytics uses real-time IoT sensor data to predict shelf life, while computer vision ensures quality monitoring. Case studies from Coles, Walmart, and Migros show AI's success in reducing waste, optimizing inventory, and boosting customer satisfaction. However, challenges include data integration, high costs, and employee training. Future trends suggest integrating AI with blockchain for better transparency and sustainability. AI can revolutionize perishable goods inventory management, making it more efficient, adaptable, and environmentally friendly.
26 September 2024 in Hacettepe Sağlık İdaresi Dergisi
Health systems assessment is critical to ensure the sustainability of health services. The interest of researchers in analysing health systems, which intensified during the pandemic, is motivated by the opportunity to identify weaknesses and propose changes. Health systems analyses provide valuable feedback for policy makers, especially in developing countries, which struggle with obstacles such as limited resources, financial constraints and deficiencies in health infrastructure. As a developing country, Türkiye has been striving to ensure equal access to health services through its health transformation programme. In the last decade, developments such as the increase in Syrian migrants and economic contraction have negatively affected this effort. This situation motivates the need to analyse the current situation in the Turkish health system and provide suggestions for improvement. The literature on the Turkish health system mostly focuses on broad health system indicators such as health service accessibility, efficiency, quality and user satisfaction. This study focuses on assessing the accessibility of health infrastructure resources. In the study, health infrastructure indicators related to the Turkish health system are evaluated on a province basis for the period from 2012 to 2021. The evaluation is based on the descriptive statistics of the dataset and the province-based health index. The findings show changing trends and gaps in access to health services with the increase in population in the period 2012-2021. Some variables (e.g. population per hospital and population per bed) show low variances, while others (e.g. population per intensive care bed) show large differences. These differences may indicate unequal development and investment in health infrastructure. For policy makers, they emphasise the need to develop policies to address these inequalities and to distribute resources equitably. Analysing the health index values shows inequalities in access to health resources in different provinces of Türkiye as a score. In particular, while some provinces such as Isparta and Bolu stand out with consistently high index scores, lower index scores in other provinces such as Şanlıurfa, Mardin, Ağrı and Şırnak indicate that inequalities in access to health services are faced. This index identifies changes and gaps in the distribution of health resources and shows policy makers priority areas for intervention.
This chapter focuses on the use of sentiment analysis in the handling of employee experience. When organisations start thinking about the experience of their employees as a factor that influences performance and turnover, sentiment analysis provides a quantitative way of measuring emotions, satisfaction and engagement of employees. This chapter attentions on several NLP techniques that can be employed in the analysis of the employee feedback which include tokenization, stop-word removal and vectorization. It also looks at how other machine learning models such as Naive Bayes, Support Vector Machines, and Long Short-Term Memory can be used to categorize emotions as positive, negative, or neutral. In addition, the problem of language, culture, and data bias is described, and the ways to solve them are also described. The future potential of the real-time emotion analysis and the use of sentiment data along with the organizational KPIs for improving the management of employees' experience is depicted.