Abstract: Forecasting electricity load is important for various purposes like planning, operation and control. Forecasts can save operating and maintenance costs, increase the reliability of power supply and delivery systems, and correct decisions for future development. This paper compares various time series methods to forecast 24 hours ahead of electricity load. The methods considered are the Holt-Winters smoothing, SARIMA Modeling, LSTM Network, Fbprophet and Tensorflow probability. The performance of each method is evaluated by using the forecasting accuracy criteria namely, the Mean Absolute Error and Root Mean Square Error. The National Renewable Energy Laboratory (NREL) residential energy consumption data are used to train the models. The results of this study show that SARIMA model is superior to the others for 24 hours ahead forecasts. Furthermore, a Bagging technique is used to make the predictions more robust. The obtained results show that by Bagging multiple time-series forecasts we can improve the robustness of the models for 24 hour ahead electricity load forecasting.
Abstract: In the last few decades, due to their advanced properties, there has been an increasing interest in hybrid composite materials. In this study, the effect of different stacking sequences of jute and carbon fabric plies on dynamic mechanical properties of composite laminates were investigated. Vacuum bagging system was used to fabricate the composite samples. Each composite laminate was reinforced with two plies of jute fabric and two plies of carbon fabric by varying the position of layers. Dynamic mechanical analyzer (DMA) was used to examine the dynamic mechanical properties of composite laminates with increasing temperature. Results showed that the composite sample, which has carbon fabric at the outer layers, has the highest storage and loss modulus. Besides, it was observed that glass transition temperature (Tg) of samples are close to each other and at about 75 °C.
Abstract: One of the major developments in machine learning in the past decade is the ensemble method, which finds highly accurate classifier by combining many moderately accurate component classifiers. In this research work, new ensemble classification methods are proposed with homogeneous ensemble classifier using bagging and heterogeneous ensemble classifier using arcing and their performances are analyzed in terms of accuracy. A Classifier ensemble is designed using Radial Basis Function (RBF) and Support Vector Machine (SVM) as base classifiers. The feasibility and the benefits of the proposed approaches are demonstrated by the means of standard datasets of intrusion detection. The main originality of the proposed approach is based on three main parts: preprocessing phase, classification phase, and combining phase. A wide range of comparative experiments is conducted for standard datasets of intrusion detection. The performance of the proposed homogeneous and heterogeneous ensemble classifiers are compared to the performance of other standard homogeneous and heterogeneous ensemble methods. The standard homogeneous ensemble methods include Error correcting output codes, Dagging and heterogeneous ensemble methods include majority voting, stacking. The proposed ensemble methods provide significant improvement of accuracy compared to individual classifiers and the proposed bagged RBF and SVM performs significantly better than ECOC and Dagging and the proposed hybrid RBF-SVM performs significantly better than voting and stacking. Also heterogeneous models exhibit better results than homogeneous models for standard datasets of intrusion detection.
Abstract: The cities of Johannesburg and Pretoria both located in the Gauteng province are separated by a distance of 58 km. The traffic queues on the Ben Schoeman freeway which connects these two cities can stretch for almost 1.5 km. Vehicle traffic congestion impacts negatively on the business and the commuter’s quality of life. The goal of this paper is to identify variables that influence the flow of traffic and to design a vehicle traffic prediction model, which will predict the traffic flow pattern in advance. The model will unable motorist to be able to make appropriate travel decisions ahead of time. The data used was collected by Mikro’s Traffic Monitoring (MTM). Multi-Layer perceptron (MLP) was used individually to construct the model and the MLP was also combined with Bagging ensemble method to training the data. The cross—validation method was used for evaluating the models. The results obtained from the techniques were compared using predictive and prediction costs. The cost was computed using combination of the loss matrix and the confusion matrix. The predicted models designed shows that the status of the traffic flow on the freeway can be predicted using the following parameters travel time, average speed, traffic volume and day of month. The implications of this work is that commuters will be able to spend less time travelling on the route and spend time with their families. The logistics industry will save more than twice what they are currently spending.
Abstract: In this study, first thermoplastic composite materials
/plates that have high ballistic impact resistance were produced. For
this purpose, the thermoplastic prepreg and the vacuum bagging
technique were used to produce a composite material. Thermoplastic
prepregs (resin-impregnated fiber) that are supplied ready to be used,
namely high-density polyethylene (HDPE) was chosen as matrix and
unidirectional glass fiber was used as reinforcement. In order to
compare the fiber configuration effect on mechanical properties,
unidirectional and biaxial prepregs were used. Then the
microstructural properties of the composites were investigated with
scanning electron microscopy (SEM) analysis. Impact properties of
the composites were examined by Charpy impact test and tensile
mechanical tests and then the effects of ultraviolet irradiation were
investigated on mechanical performance.
Abstract: This study proposes a novel recommender system that uses data mining and multi-model ensemble techniques to enhance the recommendation performance through reflecting the precise user’s preference. The proposed model consists of two steps. In the first step, this study uses logistic regression, decision trees, and artificial neural networks to predict customers who have high likelihood to purchase products in each product group. Then, this study combines the results of each predictor using the multi-model ensemble techniques such as bagging and bumping. In the second step, this study uses the market basket analysis to extract association rules for co-purchased products. Finally, the system selects customers who have high likelihood to purchase products in each product group and recommends proper products from same or different product groups to them through above two steps. We test the usability of the proposed system by using prototype and real-world transaction and profile data. In addition, we survey about user satisfaction for the recommended product list from the proposed system and the randomly selected product lists. The results also show that the proposed system may be useful in real-world online shopping store.
Abstract: Textile structures are engineered and fabricated to
meet worldwide structural applications. Nevertheless, research
varying textile structure on natural fibre as composite reinforcement
was found to be very limited. Most of the research is focusing on
short fibre and random discontinuous orientation of the reinforcement
structure. Realizing that natural fibre (NF) composite had been
widely developed to be used as synthetic fibre composite
replacement, this research attempted to examine the influence of
woven and cross-ply laminated structure towards its mechanical
performances. Laminated natural fibre composites were developed
using hand lay-up and vacuum bagging technique. Impact and
flexural strength were investigated as a function of fibre type (coir
and kenaf) and reinforcement structure (imbalanced plain woven,
0°/90° cross-ply and +45°/-45° cross-ply). Multi-level full factorial
design of experiment (DOE) and analysis of variance (ANOVA) was
employed to impart data as to how fibre type and reinforcement
structure parameters affect the mechanical properties of the
composites. This systematic experimentation has led to determination
of significant factors that predominant influences the impact and
flexural properties of the textile composites. It was proven that both
fibre type and reinforcement structure demonstrated significant
difference results. Overall results indicated that coir composite and
woven structure exhibited better impact and flexural strength. Yet,
cross-ply composite structure demonstrated better fracture resistance.
Abstract: Traffic incident has bad effect on all parts of society
so controlling road networks with enough traffic devices could help
to decrease number of accidents, so using the best method for
optimum site selection of these devices could help to implement good
monitoring system. This paper has considered here important criteria
for optimum site selection of traffic camera based on aggregation
methods such as Bagging and Dempster-Shafer concepts. In the first
step, important criteria such as annual traffic flow, distance from
critical places such as parks that need more traffic controlling were
identified for selection of important road links for traffic camera
installation, Then classification methods such as Artificial neural
network and Decision tree algorithms were employed for
classification of road links based on their importance for camera
installation. Then for improving the result of classifiers aggregation
methods such as Bagging and Dempster-Shafer theories were used.
Abstract: The study was performed to evaluate the effect of Sgirdling,
fruit thinning plus bagging with 2,4-D application, fruit
thinning plus bagging on growth and quality of wax apple fruit.
Girdling was applied three week before flowering. The 2,4-D was
sprayed at the small bud and petal fall stage. The effect of all
treatments on fruit growth was measured weekly. The physical and
biochemical quality characteristics of the fruits were recorded. The
results showed that no significant effect on number of bud among
treatments. S-girdling, 2,4-D application produced the lowest bud
drop, fruit drop compared to untreated control. Moreover, S-girdling
enhanced faster fruit growth producing the best final fruit length and
diameter than the control treatment. It was also observed that Sgirdling
greatly increased fruit set, fruit weight as well as total
soluble solid, reduced fruit crack, and titratable acidity. In
conclusion, S-girdling had a distinctive and significant effect on most
of the fruit quality characteristics assessed. Application 2,4-D was
also recommended as the industry norm to increase fruit set, and fruit
quality in wax apple.
Abstract: Ensemble learning algorithms such as AdaBoost and
Bagging have been in active research and shown improvements in
classification results for several benchmarking data sets with mainly
decision trees as their base classifiers. In this paper we experiment to
apply these Meta learning techniques with classifiers such as random
forests, neural networks and support vector machines. The data sets
are from MAGIC, a Cherenkov telescope experiment. The task is to
classify gamma signals from overwhelmingly hadron and muon
signals representing a rare class classification problem. We compare
the individual classifiers with their ensemble counterparts and
discuss the results. WEKA a wonderful tool for machine learning has
been used for making the experiments.
Abstract: The study was carried out to evaluated effect of S-gridling on fruit growth and quality of wax apple. The study was laid in Random completed block design with four replicated. Four treatment were applied as follows: S-girdling, fruit thinning plus bagging with 2,4-D sprayed, fruit thinning plus bagging and the control treatment. 2,4D was sprayed at the small bud and petal fall stage. Girdling was applied three week before flowering. The effect of all treatments on fruit growth was measured weekly. Number of flower, fruit set, fruit drop, fruit crack, and fruit quality were recorded. The result indicated that S-girdling, 2,4D application produced the lowest bud drop, fruit drop compared to untreated control. S-girdling improved faster fruit growth producing the best final fruit length and diameter compared to untreated control. S-girdling also markedly enhanced fruit set, fruit weight, and total soluble solid, reduced fruit crack, titratable acidity. On the other hand, it was noticed that with 2,4-D application also increased the fruit growth rate, improved physiological and biochemical characters of fruit than control treatment. It was concluded that S-girdling was recommended as the industry norm to increase fruit set, fruit quality in wax apple. 2,4D application had a distinctive and significant effect on most of the fruit quality characteristics assessed.
Abstract: Bagging and boosting are among the most popular re-sampling ensemble methods that generate and combine a diversity of regression models using the same learning algorithm as base-learner. Boosting algorithms are considered stronger than bagging on noise-free data. However, there are strong empirical indications that bagging is much more robust than boosting in noisy settings. For this reason, in this work we built an ensemble using an averaging methodology of bagging and boosting ensembles with 10 sub-learners in each one. We performed a comparison with simple bagging and boosting ensembles with 25 sub-learners on standard benchmark datasets and the proposed ensemble gave better accuracy.
Abstract: In this paper we designed and implemented a new
ensemble of classifiers based on a sequence of classifiers which were
specialized in regions of the training dataset where errors of its
trained homologous are concentrated. In order to separate this
regions, and to determine the aptitude of each classifier to properly
respond to a new case, it was used another set of classifiers built
hierarchically. We explored a selection based variant to combine the
base classifiers. We validated this model with different base
classifiers using 37 training datasets. It was carried out a statistical
comparison of these models with the well known Bagging and
Boosting, obtaining significantly superior results with the
hierarchical ensemble using Multilayer Perceptron as base classifier.
Therefore, we demonstrated the efficacy of the proposed ensemble,
as well as its applicability to general problems.
Abstract: Bagging and boosting are among the most popular resampling ensemble methods that generate and combine a diversity of classifiers using the same learning algorithm for the base-classifiers. Boosting algorithms are considered stronger than bagging on noisefree data. However, there are strong empirical indications that bagging is much more robust than boosting in noisy settings. For this reason, in this work we built an ensemble using a voting methodology of bagging and boosting ensembles with 10 subclassifiers in each one. We performed a comparison with simple bagging and boosting ensembles with 25 sub-classifiers, as well as other well known combining methods, on standard benchmark datasets and the proposed technique was the most accurate.
Abstract: Leo Breimans Random Forests (RF) is a recent
development in tree based classifiers and quickly proven to be one of
the most important algorithms in the machine learning literature. It
has shown robust and improved results of classifications on standard
data sets. Ensemble learning algorithms such as AdaBoost and
Bagging have been in active research and shown improvements in
classification results for several benchmarking data sets with mainly
decision trees as their base classifiers. In this paper we experiment to
apply these Meta learning techniques to the random forests. We
experiment the working of the ensembles of random forests on the
standard data sets available in UCI data sets. We compare the
original random forest algorithm with their ensemble counterparts
and discuss the results.
Abstract: A simple mobile engine-driven pneumatic paddy
collector made of locally available materials using local
manufacturing technology was designed, fabricated, and tested for
collecting and bagging of paddy dried on concrete pavement. The
pneumatic paddy collector had the following major components:
radial flat bladed type centrifugal fan, power transmission system,
bagging area, frame and the conveyance system. Results showed
significant differences on the collecting capacity, noise level, and fuel
consumption when rotational speed of the air mover shaft was varied.
Other parameters such as collecting efficiency, air velocity,
augmented cracked grain percentage, and germination rate were not
significantly affected by varying rotational speed of the air mover
shaft. The pneumatic paddy collector had a collecting efficiency of
99.33 % with a collecting capacity of 2685.00 kg/h at maximum
rotational speed of centrifugal fan shaft of about 4200 rpm. The
machine entailed an investment cost of P 62,829.25. The break-even
weight of paddy was 510,606.75 kg/yr at a collecting cost of 0.11
P/kg of paddy. Utilizing the machine for 400 hours per year
generated an income of P 23,887.73. The projected time needed to
recover cost of the machine based on 2685 kg/h collecting capacity
was 2.63 year.
Abstract: In recent years, a number of works proposing the
combination of multiple classifiers to produce a single
classification have been reported in remote sensing literature. The
resulting classifier, referred to as an ensemble classifier, is
generally found to be more accurate than any of the individual
classifiers making up the ensemble. As accuracy is the primary
concern, much of the research in the field of land cover
classification is focused on improving classification accuracy. This
study compares the performance of four ensemble approaches
(boosting, bagging, DECORATE and random subspace) with a
univariate decision tree as base classifier. Two training datasets,
one without ant noise and other with 20 percent noise was used to
judge the performance of different ensemble approaches. Results
with noise free data set suggest an improvement of about 4% in
classification accuracy with all ensemble approaches in
comparison to the results provided by univariate decision tree
classifier. Highest classification accuracy of 87.43% was achieved
by boosted decision tree. A comparison of results with noisy data
set suggests that bagging, DECORATE and random subspace
approaches works well with this data whereas the performance of
boosted decision tree degrades and a classification accuracy of
79.7% is achieved which is even lower than that is achieved (i.e.
80.02%) by using unboosted decision tree classifier.