COVID-19 has spread to more than a hundred countries worldwide since the first case reported in late 2019 in Wuhan, China. As one of the countries affected by the spread of COVID-19 cases, the local government of Malaysia has issued several policies to reduce the spread of this outbreak. One of the measures taken by the Malaysian government, namely the Movement Control Order, has been carried out since March 18, 2020. In order to provide precise information to the government so that it can take the appropriate measures, many researchers have attempted to predict and create the model for these cases to identify the number of cases each day and the peak of this pandemic. Therefore, hospitals and health workers can anticipate a surge in COVID-19 patients. In this research, confirmed, recovered, and death cases prediction was performed using the neural network as one of the machine learning methods with high accuracy. The neural network model used is the Multi-Layer Perceptron, Neural Network Auto-Regressive, and Extreme Learning Machine. The three models calculated the average percentage error (APE) values for 7 days and obtained APE values for most cases less than 10%; only 1 case in the last day of one method had an APE value of approximately 11%. Furthermore, based on the best model, then the forecast is made for the next 7 days. In conclusion, this study identified that the MLP model is the best model for 7-step ahead forecasting for confirmed, recovered, and death cases in Malaysia. However, according to the result of testing data, the ELM performs better than the MLP model.

There is an acute need to evaluate the energy consumption of buildings in response to climate change. The “occupant” factor has been largely overlooked in building energy analysis. This research aims at investigating occupancy existence in the office environment using a hybrid artificial neural network with metaheuristic algorithms for improved energy management. It proposes and compares three classification models, namely particle swarm optimization (PSO), gravitational search algorithm (GSA), and hybrid PSO-GSA in combination with the feedforward neural network (FFNN). The inputs to these models are data related to temperature, humidity, light, and carbon dioxide emissions. Two data sets are used for testing the models while the office door is open and closed. The capabilities of the optimized models are evaluated using best, average, median, and standard deviation of the mean squared error. Most of the performance metrics indicate that the FFNN-PSO-GSA model exhibits better performance compared to the other models using the two datasets. The proposed model yields a classification accuracy ranging between 98.47-98.73% using one predictor (i.e., temperature). Besides, it yields an accuracy ranging between 85.45-94.03% using temperature and CO2 predictors. It can be concluded that the FFNN combined with PSO and GSA algorithms can be a useful tool for occupancy detection modeling.

Rising sea surface height is one of the world's vital issues in marine ecosystems because it greatly affects the ecosystems as well as the socio-economic life of the surrounding environment. Pekalongan is one area in Indonesia facing the effects of this phenomenon. This problem deserves to be explored further with complex approaches. One of them is a neural network to perform forecasting more accurately. In neural networks, the time series approach can be used with Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). By adding the bidirectional method to each of these two approaches, we will find the best method to use to perform the analysis. The best results were obtained by forecasting for 960 days using Vanilla BiGRU. The results can be interpreted from multiple perspectives. The forecasting results showed a fluctuating pattern as in previous periods, so it can be said that the pattern is still quite normal, which indicates that the terminal can continue to operate normally. However, the forecasting results from this study are expected to be a reference for information for the government to prevent future dangers.

Over the past four decades, portfolio selection has been one of the most important con-cerns of researchers, project managers, project-oriented companies, and public agencies around the world. Although numerous studies have been done in this field, still there is a room for more improvement in both theory and practice. One of the yet unspoiled topics in this field is improving and balancing the efficiency of sub-portfolios while paying attention to portfolio optimization. This study employs data-mining tools to categorize projects into sub-portfolios and rank them. Multiple Criteria Decision Making (MCDM) methods are also used to weigh the criteria on which the ranking process is based. Finally, a novel multi-objective model is designed to optimize the efficiency of sub-portfolios and the gain of the main portfolio. The model is solved by NSGA II algorithm. This study introduces a hybrid framework by which project portfolio selection process can be carried out regarding strategic alignment, cost, and risk.

The World Wide Web is a great source of information, which is nowadays being widely used due to the availability of useful information changing, dynamically. However, the large number of webpages often confuses many users and it is hard for them to find information on their interests. Therefore, it is necessary to provide a system capable of guiding users towards their desired choices and services. Recommender systems search among a large collection of user interests and recommend those, which are likely to be favored the most by the user. Web usage mining was designed to function on web server records, which are included in user search results. Therefore, recommender servers use the web usage mining technique to predict users’ browsing patterns and recommend those patterns in the form of a suggestion list. In this article, a recommender system based on web usage mining phases (online and offline) was proposed. In the offline phase, the first step is to analyze user access records to identify user sessions. Next, user profiles are built using data from server records based on the frequency of access to pages, the time spent by the user on each page and the date of page view. Date is of importance since it is more possible for users to request new pages more than old ones and old pages are less probable to be viewed, as users mostly look for new information. Following the creation of user profiles, users are categorized in clusters using the Fuzzy C-means clustering algorithm and S(c) criterion based on their similarities. In the online phase, a neural network is offered to identify the suggested model while online suggestions are generated using the suggestion module for the active user. Search engines analyze suggestion lists based on rate of user interest in pages and page rank and finally suggest appropriate pages to the active user. Experiments show that the proposed method of predicting user recent requested pages has more accuracy and cover than other methods.

Cash flow models are one of the spotlights for evaluating a project. The actual data should be modeled then it could be used for the prediction process. In this paper, 996 airplane maintenance basis data are used as a database, and 119 similar data are chosen after clustering. The project is divided into 20 equal periods and first three periods are used for simulating the next point. The predicted data for each point is achieved by using of previous points from the beginning. The model is based on artificial neural network, and it is trained by three algorithms which are Genet-ic Algorithm (GA), Estimation of Distribution Algorithm (EDA), and hybrid GA-EDA method. Two dynamic ratios are used which are dividing the population into two halves, and the other is a ratio without dividing. The ratio would give a proportion to GA and EDA models in the hybrid algorithm, and then the hybrid algorithm could model the system more accurately. For each algorithm, three main errors are calculated which are mean absolute percentage error (MAPE), mean square error (MSE), and root means square error (RMSE). The best result is achieved for hybrid GA-EDA model without dividing the population and the MAPE, RMSE, and MSE values are %0.022, 28944.59 Dollars, and 837789503.79 Dollars, respectively.

Knowledge management plays an important role in modern management systems since many existing systems move towards learning organizations. Expert systems, on the other hand, are considered as the most popular techniques for adapting recent developments on knowledge management. This paper presents an empirical investigation to find important factors influencing adaptation of expert systems. The proposed study designs a questionnaire in Likert scale consists of 25 questions, distributes it among 258 people who have recently graduated from computer science and they are familiar with implementation of expert systems. Cronbach alpha is calculated as 0.730 and Kaiser-Meyer-Olkin Measure of Sampling Adequacy and Approx. Chi-Square are 0.748 and 1377.397, respectively. The study has implemented principal component analysis and the results have indicated that there were four factors influencing expert systems including systems management, intelligence systems, system analysis and specialized analysis.

Customers are believed to be the main part of any organization’s assets and customer retention as well as customer churn management are important responsibilities of organizations. In today’s competitive environment, organization must do their best to retain their existing customers since attracting new customers cost significantly more than taking care of existing ones. In this paper, we present a hybrid method based on neural network and Cox regression analysis where neural network is used for outlier data and Cox regression method is implemented for prediction of future events. The proposed model of this paper has been implemented on some data and the results are compared based on five criteria including prediction accuracy, errors’ type I and II, root mean square error and mean absolute deviation. The preliminary results indicate that the proposed model of this paper performs better than alternative methods.