The tasks of finding and selecting an accurate computational method that exists to undertake individual characteristics with various computational methods were considered difficult and would take a long completing time. The main objective of this research is to conduct a thorough study involving techniques of various computational methods that normally used in modeling and forecasting real-world problems. This paper presents the comparison results of the computational modeling methods that tested on electricity consumption data of Sarawak Energy Malaysia. The three computational methods compared in this study were Box-Jenkins technique, regression method, and artificial neural network. The models were tested on data collected from Sarawak Energy in Malaysia with regard to electricity consumption by using MATLAB software. The verification of the three methods was done using the computational statistics measurement namely the root means square error and the mean absolute percentage error. The results show that the artificial neural network was the most outperformed technique in generating the accurate prediction.

In this paper, the Adomian Decomposition Method (ADM) is employed in solving second order ordinary differential equation. Numerical algorithm was developed. The decomposition method provides a solution as an infinite series in which terms can easily be determined. It is observed that the method is practically suited for initial value problems. The method is effective and easy to implement. The results were presented in both tabular and graphical forms.

- Machine learning is a branch of artificial intelligence that is used to analyze large set of data. Machine learning approach is a statistical approach on learning more about a raw data set. When considering the existing systems in the world, there is a huge output of data which are not well analyzed. The use of machine learning techniques provide a way of analyzing a huge data set in order to find patterns and relationships among different entities which cannot be observed without advanced analyzing techniques. In this paper, the machine learning techniques that will be considered include; Box-Jenkins method, artificial neural network (ANN) technique, and Kalman technique. Each technique will be implemented using python, and the results obtained using the mentioned methods will be compared. This paper explores the application of effective machine learning to overcome challenges associated with data analysis and demonstrates how machine learning techniques have contributed and are contributing to research in machine learning.

A large proportion of online comments present on public domains are usually constructive, however a significant proportion are toxic in nature. Dataset is obtained online which are processed to remove noise from the dataset. The comments contain lot of errors which increases the number of features manifold, making the machine learning model to train the dataset by processing the dataset, in the form of transformation of raw comments before feeding it to the Classification models using a machine learning technique known as the term frequency-inverse document frequency (TF-IDF) technique. The logistic regression technique is used to train the processed dataset, which will differentiate toxic comments from non-toxic comments. The multi-headed model comprises toxicity (severetoxic, obscene, threat, insult, and identity-hate) or Non-Toxicity Evaluation, using confusion metrics for their prediction. Keywordsonline comments, toxicity, classification models, TFIDF technique, logistic regression

The purpose of this study is to develop models for controlling electricity consumption with the goal of evolving efficiency in the electricity consumption system. The models developed for the electricity consumption problem attempts to investigate the consumption pattern of individual electric appliances in a building to allow for more efficient electricity consumption. The time-based electricity consumption visualizations for appliances used in this research study is carried out to evaluate the level of efficiency in electricity consumption. This paper presents a bottom-up modelling approach for stochastic electricity consumption data profiles in households. By collecting household electricity consumption data, a model is developed for domestic electricity consumption based on daily activity profiles for individual appliances. As a means of validating the model, a statistical comparison is made between measured data collected for appliances over a period in Hamilton, New Zealand and simulated data sets from these measurements. The output of the proposed domestic load model may be designed to meet specific requirements of consumers or integrated into other models.

This study evaluates a number of power-saving measures that were applied on individual electric appliances to investigate their contributions to energy savings of the electricity network. In order to carry out the study, power-saving measures were applied on all appliances under study for certain period periods at the Universiti Malaysia Sarawak (UNIMAS). The appliances are the air conditioner (AC), computers, lightings and closed-circuit television (CCTV). As a means of validating the accuracy of models developed for electricity costs, a comparison of was done between measurements taken from the electricity network and those taken from PowerLogic PM5350 power meter (PM5350) installed for the purpose of this research. The results from model analysis show significant cost savings of 39.9%, 20.3%, 8% and 0.6% when control strategies were applied to AC, lightings, computers and CCTV

This research explores the dynamic relationship between temperature and level of building occupancy; and their effect on electricity consumption of electric appliances.It develops a model for electricity consumption based on these variables.It is important that reliable electricity consumption models are employed in finding solution to energy needs, otherwise inappropriate models may result in poor estimates for decision making.In this research, models for the daily electricity consumption for a local university in Malaysia was developed based on extraneous factors, such as temperature and level of building occupancy .As a result of developing such models, social and economic welfare will be improved.

This paper investigates related research work on electricity consumption in buildings and outlines its context relative to improving efficiency in appliance usage by customers.The aim of this research is to study the impact of applying powersaving measures on appliance usage in order to reduce electric costs.The study focuses on a review of tools and methods involved in achieving efficient electricity consumption system with respect to minimization of electric costs and reduction of electricity wastage in the system.It also conducts a survey of various literatures involving the potential impact of incorporating power-saving measures on low-power and highpower appliances to allow for more efficient use of electrical appliances.The paper provides a number of recommendations for achieving efficiency in electricity consumption, when power-saving measures are applied to appliance usage.

In this paper, the adapted time-series regression (ATSR) model is used for developing appliance energy usage profiles for a building which utilizes meter readings, and individual appliance usage using data measurements from installed power meter respectively.For this purpose, statistical models were produced for a building as well as for individual appliances.This assists in understanding the usage patterns for all types of appliances and identifies the factors that may affect the pattern.In addition, establishing a general model for a building based on different appliance use will provide more precise data than developing a model based on total consumption for the building.This will provide an insight into the contribution of each appliance on total consumption.

This study investigates the performance of regression model, Kalman filter adaptation algorithm and artificial neural network to assess their qualities for predictions. It develops predictive algorithms based on price, temperature and humidity as multiple variables affecting time-varying aspect of electricity consumption. In order to meet energy demand through the use of electricity as an energy source for daily activities in buildings such as air conditioning, lighting, computers and cooking stoves., adequate allocation of energy resources and planning should be done, including predicting for electricity consumption. The process involves collecting data from the power grid of Faculty of Computer Science and Information Technology building, Universiti Malaysia Sarawak. The forecasting techniques were tested on the data collected, and the dataset consists of electricity consumption readings, with electricity price, humidity and temperature included in the forecasting model. The performances of regression model, artificial neural network and Kalman algorithm were tested using statistical evaluation parameters, root mean squared error (RMSE) and mean absolute percentage error (MAPE); while the parameter, standard deviation, was used to check the validity of models. This study identified Kalman algorithm as the most effective method of predicting consumption data compared to regression model, and artificial neural network.

This research explores the dynamic relation between price, temperature and humidity; and its effect on electricity consumption of electric appliances. It develops prediction models for electricity consumption based on these variables. It is important that reliable methods are employed in modelling and prediction of energy needs otherwise inappropriate models and poor forecasts may occur. In this research, prediction estimates for the daily electricity consumption for a local university in Malaysia was computed using regression model, artificial neural network (ANN) and the kalman filter adaptation algorithm. The estimates of the methods were compared using performance measures based on statistical parameters obtained from identifying the difference between actual and predicted values. This research identified the kalman filter adaptation algorithm as the bests performing method in making predictions for future electricity consumption.

This paper presents a bottom-up modelling approach for stochastic production of electricity consumption profiles in households. It represents a preliminary work on individual appliance use modelling in households, as part of a bottom-up simulation to assess the impact of household consumption, and changes to consumption patterns and behavior, on the overall energy grid. By collecting household electricity consumption data, a model is developed based on daily activity profiles for individual appliances. The domestic load model obtained from simulating electricity consumption for household appliances will enable the large-scale simulation of multiple households to gain insight into individual household implications of demand-side load management strategies, as well as the combined effects on the electricity grid.

This paper investigates research work related to the modelling and simulation of household electricity consumption with a view to developing a simulation to evaluate the effectiveness of demand-side management mechanisms. The eventual aim of the research is to be able to model household consumption down to the level of individual appliance use in order to explore and assess the impact of different demand-side strategies, both in individual household consumption, and on overall grid balance. The focus of this paper is to survey relevant research on simulation of household consumption, potential demand-side strategies and their impact, and modelling techniques for residential consumption. From this review, the paper provides a number of pointers for future effort in the area of modelling the impact of demand-side management strategies and techniques.

In this paper we present modified Newton's model (MNM) to model electricity consumption data.A previous method to model electricity consumption data was done using forecasting technique (FT) and artificial neural networks (ANN).A drawback to previous techniques is that computations give less reliable results when compared to MNM.A comparative analysis is carried out for FT, ANN and MNM to investigate which of these methods is the most reliable technique.The results indicate that MNM model reduced mean absolute percentage error (MAPE) to 0.93%, while those of FT and ANN were 3.01% and 3.11%, respectively.Based on these error measures, the study shows that the three methods are highly accurate modeling techniques, but MNM was found to be the best technique when mining information.Experimental results indicate that MNM is the most accurate when compared to FT and ANN and thus has the best competitive performance level.

paper investigates the potential of applying different control measures on low power and high power appliances with the goal of evolving efficiency in electricity consumption. The research involves carrying out simulations on their power consumption readings to set up a control system. The study discovers savings on all appliances under study to be 12.8% Kw, not minding occupancy rate of the building. Air-conditioners have the greatest impact of a 6% Kw contribution on savings. This would lead to a substantial contribution when converted to pricing rates. The results from the study indicate that control measures should be extended to peak periods and power saving measures extended to more appliances.

The issue of obtaining reliable forecasting methods for electricity consumption has been widely discussed by past research work.This is due to the increased demand for electricity and as a result, the development of efficient pricing models.Several techniques have been used in past research for forecasting electricity consumption.This includes the use of forecasting, time-series technique (FTST) and artificial neural networks (ANN).This paper introduces a modified Newton's model (MNM) to forecast electricity consumption.Forecasting models are developed from historical data and predictive estimates are obtained.This research work utilizes data from Universiti Malaysia Sarawak, a public university in Malaysia, from 2009 to 2012.The variables considered in this research include electricity consumption for different months over the years.