Asian Journal of Research in Computer Science

Campus placement is a measure of students’ performance in a course. A forecasting method is proposed in this paper to predict possible campus placement of any institution. Data mining and knowledge discovery processes on academic career of students are applied. Supervised machine learning technique based classifiers are used for achieving this process. It uses an ensemble approach based voting classifier for choosing best classifier models to achieve better result over other classifiers. Experimental results have indicated 86.05% accuracy of ensemble based approach which is significantly better over other classifiers.

Restaurant industry has become one of the most profitable industries in the world where incessant long waiting time may not only lead to customers’ dissatisfactions but also facilitate loosing of customers to other competitors. In this paper, in order to determine customers’ arrival patterns and service patterns which are critical factors in determining customers’ queue length and waiting time for a given restaurant, the food serving process employed at a named International Institute Restaurant (IIR), Ibadan, Nigeria, was used as a case study. Data were collected on customers’ number, customers’ inter arrival time and service time from Monday to Friday for a week. The data were analyzed statistically using the ARENA Input Analyzer to determine the arrival patterns and service patterns of customers within five working days of the week (Monday, Tuesday, Wednesday, Thursday and Friday). The results of the data analysis revealed that the arrival times of customers who patronized the IIR on Monday and Tuesday followed a Beta distribution. Furthermore, the arrival times of customers patronizing the IIR on Wednesday and Thursday followed a Weibull distribution while that of Friday assumed an Erlang distribution. Besides, the results of the data analysis revealed that the service times at IIR on Monday and Tuesday followed a Lognormal distribution. Beta, Lognormal and Weibull distributions were recorded in respect of service times characterizing the IIR on Wednesday, Thursday and Friday, respectively.

COVID-19 disease came to earth in December 2019 in Wuhan. It is increasing exponentially throughout the world and affected an enormous number of human beings. The World Health Organization (WHO) on March 11, 2020 declared COVID-19 was characterized as “Pandemic”. Clinical Doctors have been working on it 24 hours in the entire world. These doctors are testing whether the particular human has been affected with the disease using testing kit and other related process. Researchers have been working day-night for developing vaccine for the disease. Since the rate of affected people is so high, it is difficult for clinical doctors to check such a large number of coronavirus detected humans within reasonable time.

This paper attempts to use Machine Learning Approach to build up model which will help clinical doctors for verification of disease within short period of time and also the paper attempts to predict growth of the disease in near future in the world. Two models were used for achieving this purpose- One is based on Convolutional Neural Network model where as another one consists of Convolutional Neural Network and Recurrent Neural Network. These two models are evaluated and compared for verifying the predicted result with respect to the original one. Experimental results indicate that the combined CNN-LSTM approach outperforms well over the other model.

The increasing rate of hospital admission led to increased volume of both municipal and biohazard wastes. The fourth industrial revolution opens up new perspectives to improve the conventional processes of hospitals and other institutions of health care systems. The application of new technologies integrated into the solutions of Industry 4.0 makes it possible to improve the efficiency of processes of hospitals not only in the field of medicine, but also in the field of other services, like logistics and supply chain. Within the frame of this article the authors are focusing on the development of new operation strategies for biohazard waste collection in hospitals using smart bins and autonomous waste collection vehicles/robots. The literature review helps to identify research gaps in the field of biohazard waste collection in hospitals. After that, the article describes the mathematical model of the waste collection system including smart bins, autonomous collection vehicles, users (patients and health professionals). The mathematical model has a time-based objective function, while time-, capacity- and safety-related constraints are also taken into consideration. The mathematical problem is an NP-hard problem; therefore, we use the non-linear regression and evolutive options of Excel Solver to find the optimal solution. The numerical scenarios validate the model and show the advantages of using new technologies in health care institutes, like hospitals.

The problems of energy usage wastage, conservation and optimization has always been there. Most significantly in third world countries with a high level of improper energy resource management. The world has become a global village and tending toward the transformational use of technology assets and materials to optimize the usage of renewable energy; in this case the Internet of things. Renewable Energy are actually energy source that are continually replenished naturally by nature. The internet of things are interconnected system through which hardware system device are given instruction for instance to reduce energy wastage; the potential for further reduction of fossil fuel usage and wasted energy is become cosmic. Optimization here is the appropriate management of the energy generated by energy sources through use of internet of things. The modern technology such as the internet of things is to help harness and improve in management by switching between renewable energy sources alongside the enhancement in making effective energy source at given time and energy consuming device.