Asian Journal of Research in Computer Science

Most industries around the globe make use of image processing to improve their productions. On the other hand Big Data Processing is a big dataset; this required fast method to processing irrespective of Generic nature, therefore Clarification of heterogeneous images can improve the integrity of any system design. To avoid waste of time and energy, it is necessary to classify images. Big Data Processing for Generic Clarification of heterogeneous images provides fast, accurate and objectives results. In this study, the researchers classified into three category using resnet50 techniques for training dataset images. The outcome of the research is analyzing these techniques and comparison analysis on different existing image data sets as pre-trained data and test data as sample images for decision making based on their limitations and strengths.

This research presents the epidemiological evaluation and statistical analysis COVID-19 Pandemic in Nigeria after three months of its first incidence. The aim is to assess the performance of the medical professionals, Nigerian Center for Disease and Control (NCDC), the Governments and the general public respectively in the fight against COVID-19 in the last three months of the first incidence case. This was done using the data collected from the NCDC and analyzed using the Microsoft BI analyzer. From the evaluation, it was observed that after three months, a total number of 8077 cases have been recorded. Of this, 68.5% are active cases, receiving treatment in the hospital, 28.6% have recovered while 2.9% have died with majority of them over 50 years in age and have cardiac related cases before the virus struck. The implication of this result shows that the health care professionals and NCDC are doing their best having recorded a very low death rate so far compared to the total recorded cases. However the government needs to support and properly equip the hospitals with enough health care resources to help optimize patient response to treatment. Finally it was observed that despite the low death rate recorded, that the rate of increased new cases is alarming. The implication is that the general public is not very supportive in this fight of COVID-19. Hence more public awareness and recommended to educate the public on the safety measures required to prevent the spread of this virus and ensure public safety.

Supervised learning in Spiking Neural Network (SNN) is a hotbed for researchers due to the advantages temporal coded networks provide over that of rate-coded networks with respect to efficiency in information processing and transfer rates. Supervised learning in rate-coded networks though well established, it is difficult to directly apply such models to SNN due to difference in information coding schemes. In this paper, we seek to exploit the advantages of spiking neural networks for spike sequence learning in order to establish two (2) models; batch and sequential learning models for solving data classification tasks. The models are built using the least squares approach leveraging on its approximation abilities. The first set of experiments are on spike
sequence learning in which an extensive evaluation of the model is performed using different inputoutput firing rates and learning periods. Results from these experiments show that the proposed model for spike sequence learning produced better performance than some existing models derived for spike sequence learning, particularly, at higher learning periods. The proposed models for data classification are also tested on some selected benchmark datasets most of which had imbalance class distributions and also on real-world road condition datasets for anomaly classification collected by the authors as part of a larger study. While the proposed models generalised very well to all datasets including those with the class imbalance problem where F1
and Recall values above 0.90 were recorded, some well-know machine learning algorithms applied to the datasets yielded lower F1 and Recall values and in some cases recorded 0.0 Recall.

The era of coronavirus has called for sustained social distancing measures to minimize the spread of the viral disease. Healthcare establishments are reducing the size of their working staff; while others are running their outfits base on shift work in other to ensure protocols for social distancing.  Inherent in social distancing protocol is the potential for generating waiting lines at service delivery points. Healthcare centres in many countries are already inundated with loads of patient’s attendance on daily basis for treatment off mild to severe ailments. COVID-19 has added a further burden on the already frail health systems. Whiles visits are increasing, social distancing measures are to be ensured. Quick service delivery which is an indispensable need of patients visiting hospitals for treatment is shortened. The occurrence of waiting line, an impediment to healthcare provision has become commonplace in most healthcare centres in Ghana in particular. In addition to loss of financial gains, delay and unsatisfactory healthcare could lead to loss of lives. Health units are dealing with the effective management of staff schedules to curtail the impact of COVID-19 and at the same time cover up capacity to meet the added health care delivery demands. Accordingly, efforts to reduce time spent in waiting to receive medical attention is crucial. In this paper we study the queue situation at a case Outpatient Department (OPD) by applying query theory and offer recommendations for queue management. The study was conducted in the month of May 2020. We present also, an approach to determine the optimal number of service windows required to reduce the time spent waiting for healthcare attention. Numerical analysis of the queuing situation at the case department is given also, drawing from relevant equations from queuing theory.

The approach presented in this article aims at transition between two systems of counting binary and ternary I propose to use ternary math principle in coding the signal Instead of using duos of numbers 01 I propose to use triplets (1,0-1) and make a transition from binary to ternary so that the binary code is converted to ternary and vice versa That same principle can be used for building microcircuits where logical elements are placed in a 3 D space instead of a layer.