Knowledge discovery helps mitigate the shortcomings of classical machine learning, especially those so-called imbalanced, high-dimensional, and noisy data challenges. Adaptive combination of multiple models, voting and other data fusion strategies, and the incorporation of other disparate information fusion methods characterize ensemble learning, which addresses the improvement of a predictive model’s accuracy, stability, and generalization. This paper provides a summary of the important approaches to ensemble learning and their real-world uses, emphasizing challenges and opportunities for future work. This paper also discusses how ensemble learning integrates with emergent areas such as deep learning and reinforcement learning. This paper also describes the most important machine learning methods for predicting heart disease, which include decision trees, support vector machines, artificial neural networks, Naïve Bayes, random forest, and K-nearest neighbors.
1 January 2018 in International Journal of Advanced Computer Science and Applications
Blended learning has emerged as one of the solutions to address the various needs of Higher Education Institutions around the world. Blended Learning is the combination of traditional classroom and online endeavour. It provides advantages of both face to face learning and e-learning. The main purpose of this study is to assess the adaptation level of blended learning in teaching process at Higher Education Institutions. This study carried out mixed method approach by using explanatory sequential model. Teachers of general public universities were included as the sample for this study. Questionnaire and interview techniques were used as data gathering tools. The main findings of this study showed that teachers have a positive perception for technology usage in teaching process. Most of the teachers possessed expertise in the use of different software and equipped with internet skills. The study concluded that in blended learning implementation, universities are still at awareness level and a lot of efforts are required for effective implementation of blended learning. It is recommended that the universities’ administration should provide an extra computing infrastructure (e.g. servers, bandwidth, and storage capacity) to run the courses in blended format. We recommend that in strategic plan of the universities the blended learning should be well defined and highlighted.
Customer satisfaction plays very important role in online shopping. There is always a drought on what actually drives the customer satisfaction in online shopping. Many studies have been done to understand the customer satisfaction in online shopping, however, as of now, there is no study found that have exactly define the user satisfaction through trust and security. These are the two factors that have been identified as a major barrier in customer's satisfaction in online shopping, especially in the business to consumer B2C e-commerce mode. We also consider the cybercrime as another major barrier to achieve the customer satisfaction in online shopping. Although numerous studies have been conducted which related to the area, there is no satisfactory solution for trust and security in online shopping has been attained, therefore, in order to address this issue, we proposed a research framework that emphasize on trust and security mechanisms to achieve customer's satisfaction. We also show the negative impact of cybercrime on customer trust and satisfaction. In this paper we present the framework that has been designed based on five essential current issues to achieve customer satisfaction in online shopping, which directly impacted the customer's trust. The essential current issues namely are financial security, personal security, system security, privacy and cybercrime.
Aluminum nanoparticles are suitable for wiring power grids, such as local power distribution and overhead power transmission lines, because they exhibit high conductivity. These nanoparticles are also among the most utilized materials in electrical field applications. Thus, the present study investigated the impact of magnetic field on 3D crossflow in the streamwise direction with the impacts of Dufour and Soret. In addition, the effects of activation energy and chemical reaction were incorporated. The viscosity and thermal conductivity of nanofluids were premeditated by KKL correlation. Prominent PDEs (Partial Differential Equations) were converted into highly nonlinear ODEs (Ordinary Differential Equations) using the proper similarity technique and then analyzed numerically with the aid of the built-in bvp4c solver in MATLAB. The impact of diverse important variables on temperature and velocity was graphically examined. Additionally, the influences of pertaining parameters on the drag force coefficient, Nusselt number, and Sherwood number were investigated. Inspections revealed that the mass transfer rate decreases, while the heat transport increases with increasing values of the Soret factor. However, the Nusselt and Sherwood numbers validate the differing trend for rising quantities of the Dufour factor.
Jamshoro is considered as the city of Education in Pakistan. It is comprised of three big universities. Where students from all over Pakistan especially Sindh Province are getting the knowledge and education. There are different types of problems which faced by the students at different education level. The focus of this paper is to elaborate the problems that are faced by the females Postgraduate students in the universities of Jamshoro cities. This research will be conducted through observation and quantitative data collection method. The pilot study has been conducted to calculate the reliability of items.
10 March 2018 in International Journal of Engineering & Technology
Representation of Pre-RST information is very useful using visualized elements for realization of benefits of requirement traceability. This improves the practitioner motivation to maintain Pre-RST information during life cycle processes. Few researchers proposed visualization for Post-RST due to which many benefits of requirement traceability cannot be realized. This paper proposed an improved visualization representing Pre-RST information that demonstrates various benefit of requirement traceability. In order to evaluate empirically, an experiment is conducted and textual representation of traceability information is obtained. In order to strengthen our claim a survey is conducted to compare textual representation of traceability information with proposed visualization and results are compiled.