No abstract

The personalized health care service utilizes the relational patient data and big data analytics to tailor the medication recommendations. However, most of the health care data are in unstructured form and it consumes a lot of time and effort to pull them into relational form. This study proposes a novel data lake architecture to reduce the data ingestion time and improve the precision of healthcare analytics. It also removes the data silos and enhances the analytics by allowing the connectivity to the third-party data providers (such as clinical lab results, chemist, insurance company,etc.). The data lake architecture uses the Hadoop Distributed File System (HDFS) to provide the storage for both structured and unstructured data. This study uses K-means clustering algorithm to find the patient clusters with similar health conditions. Subsequently, it employs a support vector machine to find the most successful healthcare recommendations for the each cluster. Our experiment results demonstrate the ability of data lake to reduce the time for ingesting data from various data vendors regardless of its format. Moreover, it is evident that the data lake poses the potential to generate clusters of patients more precisely than the existing approaches. It is obvious that the data lake provides an unified storage location for the data in its native format. It can also improve the personalized healthcare medication recommendations by removing the data silos.

Due to the popularity of Web-based applications, various developers have provided an abundance of Web services with similar functionality. Such similarity makes it challenging for users to discover, select, and recommend appropriate Web services for the service-oriented systems. Quality of Service (QoS) has become a vital criterion for service discovery, selection, and recommendation. Unfortunately, service registries cannot ensure the validity of the available quality values of the Web services provided online. Consequently, predicting the Web services' QoS values has become a vital way to find the most appropriate services. In this paper, we propose a novel methodology for predicting Web service QoS using source code metrics. The core component is aggregating software metrics using inequality distribution from micro level of individual class to the macro level of the entire Web service. We used correlation between QoS and software metrics to train the learning machine. We validate and evaluate our approach using three sets of software quality metrics. Our results show that the proposed methodology can help improve the efficiency for the prediction of QoS properties using its source code metrics.

In service computing, the same target functions can be achieved by multiple Web services from different providers. Due to the functional similarities, the client needs to consider the non-functional criteria. However, Quality of Service provided by the developers suffers scarcity and lack of reliability. In addition, the reputation of the service providers is an important factor, especially those with little experience, to select a service. Most of the previous studies were focused on the user's feedbacks for justifying the selection. Unfortunately, not all the users provide the feedback unless they had extremely good or bad experience with the service. In this vision paper, we propose a novel architecture for the web service discovery and selection. The core component is a machine learning based methodology to predict the QoS properties using source code metrics. The credibility value and previous usage count are used to determine the reputation of the service.

Web Services are the most emerging distributed applications published in the public registries. Web service are can be discovered by both functional properties and non functional properties. Due to the rapid Web development, there are number of functionally similar Web Services published by different vendors. The functional property based web service discovery is cannot be done with accuracy. So client can find the best Web Services by taking the non-functional criteria such as Quality of Service (QoS). However, most of clients are not experienced enough to acquire the best selection of Web Service based on its described QoS properties. In this paper we are proposing a client request message structure and broker architecture to find the best Web Service. First the broker will get Web Service client's requirement message along with QoS criteria, and then it will retrieve the functionally similar web service. The broker will use an efficient mechanism to rank the Web Services based on the client's message as well as the QoS properties which being confirmed by the broker architecture, If any tie situation happens in the ranking of the web service we will use the previous usage history of the web service to select the best web service which is matching with the client's request message.