Search

Dengue is a mosquito-borne viral infection transmitted by the bite of an infected vector. In the resent past dengue cases are being increased uncontrollably across the world. Due to the dynamic nature of the dengue virus, a specific treatment has not been discovered yet and controlling can be considered as the only mitigating techniques. Presently, in Sri Lanka prevention procedure of dengue is based on reactive systems rather than proactive systems. Further, there is a need of general optimum infrastructure or mechanism, in which all the sectors who are contributing for the epidemic controlling can collaborate together under efficient action plan. Pro Active Dengue Management System (ProDMS) is a multi-platform GIS based distributed system for monitoring, forecasting and controlling the dengue epidemics. ProDMS consists of two main components, which can be introduced as Mobile Application and Web base System. In this study, we present suitable dynamic programming technics to develop some features of ProDMS android mobile application. The features are developed based on the dynamic programming methods rather than using existing static approaches. The features, are Statistical Charts, education packages, Alerting and Images of mosquito breeding places. All features are designed through dynamically changing menus with back end control. Content pages and statistical charts are designed in a way that they can function according to real time data and dynamic ongoing changes. The developed application uses JSON files as a medium to distribute the structured data among its components. Further, ensuring the confidentiality and the security of the data, transferring JSON files are embedded within user level securely measure when they transfer through web application. Concerning the sustainability of the system like ProDMS, all the software are designed so that it can be administrated and updated by privilege users who has basic computer literacy. [ABSTRACT FROM AUTHOR]

Most of the current portfolio optimization techniques depend only on historical data. However, in a reality, decisions may be changed due to the behavior of investors. Investor’s views may depend on various external factors, such as current market stability, company performances and many factors related to government policies. However, modeling the investor’s behavior pattern is too complex a process due to uncertainty of such patterns. This study is an attempt to discuss how the optimum allocation changes with respect to investor’s views, with a quantitative approach. Therefore, present study focuses to find the optimum asset allocation based on both historical data and investors’ view combined. The raw data set is weighted using an information matrix which represents the investors view, and then the weighted dataset is used to find the optimum asset allocation. The processes are done with the different sets of weights and obtain the corresponding optimum allocations. The optimum allocation is changed with respect to the weightage. For current study, five companies are selected. The historical return of closing price of the chosen companies is weighted using information matrix. Sensitivity of the assets allocation is discussed with respect to varying information. For this purpose, the information matrix is changed and thus different weightages are obtained to get different allocations. [ABSTRACT FROM AUTHOR]