Electronic Patient Health Record (EPHR) is an individual official patient health document shared among multiple facilities and agencies. However, most existing EPHR systems are dependent on keyboard and mouse only and do not support human speech interaction. This study, therefore, developed an improved EPHR management system, characterized by human speech interaction. Oral interview was conducted, the information acquired was used to design the improved system whose components are speech recognition architecture, speech recognition algorithm and voice command algorithm. The improved EPHR system database was developed using Microsoft SQL server 2016, Legacy ActiveX Data Objects (ADO.net) with Object Relational Mapping. Microsoft speech library was used for the speech recognition module. The improved EPHR system was implemented using C# programming language (.NET 4.5) and Visual Studio 2017. The performance of the improved EPHR system with speech recognition was evaluated with 50, 100, 150 and 200 words using correctness, accuracy and Word Error Rate (WER). The performance of the improved EPHR system yielded correctness, accuracy and WER values of (96, 96 and 4.0%), (96, 95 and 4.0%), (95, 95 and 5.0%) and (93, 94 and 6.0%) for 50, 100, 150 and 200 words respectively. This study developed an improved EPHR management system with speech recognition and voice command which can improve user interactivity and help in disabilities or hands-free environment.
In this paper, a two- stage stochastic fully fuzzy linear programming is developed for a management problem in terms of water resources allocations to illustrate the applicability of a proposed approach. A proposed approach converts the problem into a triple- objective problem and then a weighting method is utilized for solving it. The advantage of the approach is to generate a set of solutions for water resources planning which help the decision maker to make tradeoffs between the efficiency of economic and the risk violation of the constrains. A case study is given for illustration.
World requires to evolve itself with a stringent personnel identification system due to increase in the number of assets and expansion in the number of stakeholders involved in their maintenance. This is constantly challenged by the newer threats. The system requires being high on quality factors such as availability, performance, robustness, durability with negligible downsides such as cost, partialness in its perusal. To ensure these high standards, society has been making way for biometric driven security schemes that are able to replicate the expected quality norms. But the existing biometric systems need to be more convergent to the customization that is oriented to the end user. Therefore, this paper intends to bring system and end user to a same platform of contribution. This is achieved when end user customizes the system as per his biometric requirement and the system can refer the user in case it is unable to exactly identify biometric traits during user authentication. Additionally, system ensures the information compliance to target audience along with a constant reporting culture as a minimum standard. This would introduce high reliability and maximum functionality to the technological ecosystem of the security world.
A linear differential equation with polynomial coefficients is studied. In the preceding study given in J. Adv. Math. Comput. Sci. 2018; 28 (3) 1-15, the equation is expressed in terms of blocks of classified terms, and the full solutions near the origin are presented for the differential equations of the second order and with two blocks of classified terms. In the present study, it is shown that the solutions near infinity are easily obtained and the asymptotic behaviors are discussed with the aid of a theorem given in the preceding paper.
World Wide Web has become a huge collection of documents and the amount of documents available is increasing on a daily basis. How to correctly classify the vast documents into a particular category and locate any document of interest easily has become a challenge researchers have been trying to solve for decades and different researchers have attempted different algorithms using different platform to achieve this aim. In this paper, a University web site was used as a case study and a machine learning workbench called WEKA (Waikato Environment for Knowledge Analysis) which provides a general-purpose environment for automatic classification, regression, clustering and feature selection was used as a machine learning platform. Running Naïve Bayes with 10-fold cross validation on the selected web data gives a 77% correctly classified instances in zero second with relative absolute error of 68.9937%. This shows the ability of Naïve Bayes algorithm to accurately classify vast amount of web document in a short time.