Search results

The amount of features in handwritten digit data is often very large due to the different aspects in personal handwriting, leading to high-dimensional data. Therefore, the employment of a feature selection algorithm becomes crucial for successful classification modeling, because the inclusion of irrelevant or redundant features can mislead the modeling algorithms, resulting in overfitting and decrease in efficiency.

The minimum redundancy and maximum relevance (mRMR) and the recursive feature elimination (RFE) are two frequently used feature selection algorithms. While mRMR is capable of identifying a subset of features that are highly relevant to the targeted classification variable, mRMR still carries the weakness of capturing redundant features along with the algorithm. On the other hand, RFE is flawed by the fact that those features selected by RFE are not ranked by importance, albeit RFE can effectively eliminate the less important features and exclude redundant features.

The hybrid method was exemplified in a binary classification between digits “4” and “9” and between digits “6” and “8” from a multiple features dataset. The result showed that the hybrid mRMR +  support vector machine recursive feature elimination (SVMRFE) is better than both the sole support vector machine (SVM) and mRMR.

In view of the respective strength and deficiency mRMR and RFE, this study combined both these methods and used an SVM as the underlying classifier anticipating the mRMR to make an excellent complement to the SVMRFE.

The present research work is carried out for determining haemoprotozoan diseases in cattle and breast cancer diseases in humans at early stage. The combination of LeNet and bidirectional long short-term memory (Bi-LSTM) model is used for the classification of heamoprotazoan samples into three classes such as theileriosis, babesiosis and anaplasmosis. Also, BreaKHis dataset image samples are classified into two major classes as malignant and benign. The hyperparameter optimization is used for selecting the prominent features. The main objective of this approach is to overcome the manual identification and classification of samples into different haemoprotozoan diseases in cattle. The traditional laboratory approach of identification is time-consuming and requires human expertise. The proposed methodology will help to identify and classify the heamoprotozoan disease in early stage without much of human involvement.

LeNet-based Bi-LSTM model is used for the classification of pathology images into babesiosis, anaplasmosis, theileriosis and breast images classified into malignant or benign. An optimization-based super pixel clustering algorithm is used for segmentation once the normalization of histopathology images is conducted. The edge information in the normalized images is considered for identifying the irregular shape regions of images, which are structurally meaningful. Also, it is compared with another segmentation approach circular Hough Transform (CHT). The CHT is used to separate the nuclei from non-nuclei. The Canny edge detection and gaussian filter is used for extracting the edges before sending to CHT.

The existing methods such as artificial neural network (ANN), convolution neural network (CNN), recurrent neural network (RNN), LSTM and Bi-LSTM model have been compared with the proposed hyperparameter optimization approach with LeNET and Bi-LSTM. The results obtained by the proposed hyperparameter optimization-Bi-LSTM model showed the accuracy of 98.99% when compared to existing models like Ensemble of Deep Learning Models of 95.29% and Modified ReliefF Algorithm of 95.94%.

In contrast to earlier research done using Modified ReliefF, the suggested LeNet with Bi-LSTM model, there is an improvement in accuracy, precision and F-score significantly. The real time data set is used for the heamoprotozoan disease samples. Also, for anaplasmosis and babesiosis, the second set of datasets were used which are coloured datasets obtained by adding a chemical acetone and stain.

In order to prevent accidents during driving, driver drowsiness detection systems have become a hot topic for researchers. There are various types of features that can be used to detect drowsiness. Detection can be done by utilizing behavioral data, physiological measurements and vehicle-based data. The existing deep convolutional neural network (CNN) models-based ensemble approach analyzed the behavioral data comprises eye or face or head movement captured by using a camera images or videos. However, the developed model suffered from the limitation of high computational cost because of the application of approximately 140 million parameters.

The proposed model uses significant feature parameters from the feature extraction process such as ReliefF, Infinite, Correlation, Term Variance are used for feature selection. The features that are selected are undergone for classification using ensemble classifier.

The output of these models is classified into non-drowsiness or drowsiness categories.

In this research work higher end camera are required to collect videos as it is cost-effective. Therefore, researches are encouraged to use the existing datasets.

This paper overcomes the earlier approach. The developed model used complex deep learning models on small dataset which would also extract additional features, thereby provided a more satisfying result.

Drowsiness can be detected at the earliest using ensemble model which restricts the number of accidents.

Water pipes degrade over time for a variety of pipe-related, soil-related, operational, and environmental factors. Hence, municipalities are necessitated to implement effective maintenance and rehabilitation strategies for water pipes based on reliable deterioration models and cost-effective inspection programs. In the light of foregoing, the paramount objective of this research study is to develop condition assessment and deterioration prediction models for saltwater pipes in Hong Kong.

As a perquisite to the development of condition assessment models, spherical fuzzy analytic hierarchy process (SFAHP) is harnessed to analyze the relative importance weights of deterioration factors. Afterward, the relative importance weights of deterioration factors coupled with their effective values are leveraged using the measurement of alternatives and ranking according to the compromise solution (MARCOS) algorithm to analyze the performance condition of water pipes. A condition rating system is then designed counting on the generalized entropy-based probabilistic fuzzy C means (GEPFCM) algorithm. A set of fourth order multiple regression functions are constructed to capture the degradation trends in condition of pipelines overtime covering their disparate characteristics.

Analytical results demonstrated that the top five influential deterioration factors comprise age, material, traffic, soil corrosivity and material. In addition, it was derived that developed deterioration models accomplished correlation coefficient, mean absolute error and root mean squared error of 0.8, 1.33 and 1.39, respectively.

It can be argued that generated deterioration models can assist municipalities in formulating accurate and cost-effective maintenance, repair and rehabilitation programs.

Indian Premier League (IPL) is one of the more popular cricket world tournaments, and its financial is increasing each season, its viewership has increased markedly and the betting market for IPL is growing significantly every year. With cricket being a very dynamic game, bettors and bookies are incentivised to bet on the match results because it is a game that changes ball-by-ball. This paper investigates machine learning technology to deal with the problem of predicting cricket match results based on historical match data of the IPL. Influential features of the dataset have been identified using filter-based methods including Correlation-based Feature Selection, Information Gain (IG), ReliefF and Wrapper. More importantly, machine learning techniques including Naïve Bayes, Random Forest, K-Nearest Neighbour (KNN) and Model Trees (classification via regression) have been adopted to generate predictive models from distinctive feature sets derived by the filter-based methods. Two featured subsets were formulated, one based on home team advantage and other based on Toss decision. Selected machine learning techniques were applied on both feature sets to determine a predictive model. Experimental tests show that tree-based models particularly Random Forest performed better in terms of accuracy, precision and recall metrics when compared to probabilistic and statistical models. However, on the Toss featured subset, none of the considered machine learning algorithms performed well in producing accurate predictive models.

Organizations in many domains generate a considerable amount of heterogeneous data every day. Such data can be processed to enhance these organizations’ decisions in real time. However, storing and processing large and varied datasets (known as big data) is challenging to do in real time. In machine learning, streaming feature selection has always been considered a superior technique for selecting the relevant subset features from highly dimensional data and thus reducing learning complexity. In the relevant literature, streaming feature selection refers to the features that arrive consecutively over time; despite a lack of exact figure on the number of features, numbers of instances are well-established. Many scholars in the field have proposed streaming-feature-selection algorithms in attempts to find the proper solution to this problem. This paper presents an exhaustive and methodological introduction of these techniques. This study provides a review of the traditional feature-selection algorithms and then scrutinizes the current algorithms that use streaming feature selection to determine their strengths and weaknesses. The survey also sheds light on the ongoing challenges in big-data research.

The purpose of this study was to build and develop mathematical models correlating ambient conditions and electrical energy to the coefficient of performance (COP) of an air-source heat pump (ASHP) water heater. This study also aimed to design a simulation application to compute the COP under different heating up scenarios, and to calculate the mean significant difference under the specified scenarios by using a statistical method.

A data acquisition system was designed with respect to the required sensors and data loggers on the basis of the experimental setup. The two critical scenarios (with hot water draws and without hot water draws) during the heating up cycles were analyzed. Both mathematical models and the simulation application were developed using the analyzed data.

The predictors showed a direct linear relationship to the COP under the no successive hot water draws scenario, while they exhibited a linear relationship with a negative gradient to the COP under the simultaneous draws scenario. Both scenarios showed the ambient conditions to be the primary factor, and the weight of importance of the contribution to the COP was five times more in the scenario of simultaneous hot water draws than in the other scenario. The average COP of the ASHP water heater was better during a heating cycle with simultaneous hot water draws but demonstrated no mean significant difference from the other scenario.

There was a need to include other prediction parameters such as air speed, difference in condenser temperature and difference in compressor temperature, which could help improve model accuracy. However, these were excluded because of insufficient funding for the purchase of additional temperature sensors and an air speed transducer.

The research was conducted in a normal middle-income family home, and all the results were obtained from the collected data from the data acquisition system. Moreover, the experiment was very feasible because the conduction of the study did not interfere with the activities of the house, as occupants were able to carry out their activities as usual.

This paper attempts to justify the system efficiency under different heating up scenarios. Based on the mathematical model, the performance of the system could be determined all year round and the payback period could be easily evaluated. Finally, from the study, homeowners could see the value of the efficiency of the technology, as they could easily compute its performance on the basis of the ambient conditions at their location.

This is the first research on the mathematical modeling of the COP of an ASHP water heater using ambient conditions and electrical energy as the predictors and by using surface fitting multi-linear regression. Further, the novelty is the design of the simulation application for a Simulink environment to compute the performance from real-time data.

This paper aims to show that a simplified surface fitting model can be efficient in determining the energy consumption during milk cooling by an on-farm direct expansion bulk milk cooler (DXBMC). The study reveals that milk volume and the temperature gradient between the room and the final milk temperature can effectively be used for predicting the energy consumption within 95% confidence bounds.

A data acquisition system comprised a Landis and Gyr E650 power meter, TMC6-HE temperature sensors, and HOBO UX120-006M 4-channel analog data logger was designed and built for monitoring of the DXBMC. The room temperature where the DXBMC is housed was measured using a TMC6-HE temperature sensor, connected to a Hobo UX120-006M four-channel analog data logger which was configured to log at one-minute intervals. The electrical energy consumed by the DXBMC was measured using a Landis and Gyr E650 meter while the volume of milk was extracted from on the farm records.

The results showed that the developed model can predict the electrical energy consumption of the DXBMC within an acceptable accuracy since 80% of the variation in the electrical energy consumption by the DXBMC was explained by the mathematical model. Also, milk volume and the temperature gradient between the room and final milk temperature in the BMC are primary and secondary contributors, respectively, to electrical energy consumption by the DXBMC. Based on the system that has been monitored the findings reveal that the DXBMC was operating within the expected efficiency level as evidenced by the optimized electrical energy consumption (EEC) closely mirroring the modelled EEC with a determination coefficient of 0.95.

Only one system was monitored due to unavailability of funding to deploy several data acquisition systems across the country. The milk blending temperatures, effects of the insulation of the DXBMC, were not taken into account in this study.

The developed model is simple to use, cost effective and can be applied in real-time on the dairy farm which will enable the farmer to quickly identify an increase in the cooling energy per unit of milk cooled.

The developed easy to use model can be used by dairy farmers on similar on-farm DXBMC; hence, they can devise ways to manage their energy consumption on the farm during the cooling of milk and foster some energy efficiency initiatives.

The implementation of the developed model can be useful to dairy farmers in South Africa. Through energy optimization, the maintenance of the DXBMC can be determined and scheduled accordingly.

On-time fault diagnosis in electrical machines is a critical issue, as it can prevent the development of fault and also reduce the repairing time and cost. In brushless synchronous generators, the significance of the fault diagnosis is even more because they are widely used to generate electrical power all around the world. Therefore, this study aims to propose a fault detection approach for the brushless synchronous generator. In this approach, a novel extension of Relief feature selection method is developed.

In this paper, by taking the advantages of the finite element method (FEM), a brushless synchronous machine is modeled to evaluate the machine performance under two conditions. These conditions include the normal condition of the machine and one diode open-circuit of the rotating rectifier. Therefore, the harmonic behavior of the terminal voltage of the machine is obtained under these situations. Then, the harmonic components are ranked by using the extension of Relief to extract the most appropriate components for fault detection. Therefore, a fault detection approach is proposed based on the ranked harmonic components and support vector machine classifier.

The proposed diagnosis approach is verified by using an experimental test. Results show that by this approach open-circuit fault on the diode rectifier can effectively be detected by the accuracy of 98.5% and by using five harmonic components of the terminal voltage [1].

In this paper, a novel feature selection method is proposed to select the most effective FFT components based on an extension of Relief method, and besides, FEM modeling of a brushless synchronous generator for normal and one diode open-circuit fault.

As drones are rapidly transforming tasks such as mapping and surveying, safety inspection and progress monitoring, human operators continue to play a critical role in ensuring safe drone missions in compliance with safety regulations and standard operating procedures. Research shows that operator's stress and fatigue are leading causes of drone accidents. Building upon the authors’ past work, this study presents a systematic approach to predicting impending drone accidents using data that capture the drone operator's physiological state preceding the accident.

The authors collect physiological data from 25 participants in real-world and virtual reality flight experiments to design a feedforward neural network (FNN) with back propagation. Four time series signals, namely electrodermal activity (EDA), skin temperature (ST), electrocardiogram (ECG) and heart rate (HR), are selected, filtered for noise and used to extract 92 time- and frequency-domain features. The FNN is trained with data from a window of length t = 3…8 s to predict accidents in the next p = 3…8 s.

Analysis of model performance in all 36 combinations of analysis window (t) and prediction horizon (p) combinations reveals that the FNN trained with 8 s of physiological signal (i.e. t = 8) to predict drone accidents in the next 6 s (i.e. p = 6) achieved the highest F1-score of 0.81 and AP of 0.71 after feature selection and data balancing.

The safety and integrity of collaborative human–machine systems (e.g. remotely operated drones) rely on not only the attributes of the human operator or the machinery but also how one perceives the other and adopts to the evolving nature of the operational environment. This study is a first systematic attempt at objective prediction of potential drone accident events from operator's physiological data in (near-) real time. Findings will lay the foundation for creating automated intervention systems for drone operations, ultimately leading to safer jobsites.