Wireless Sensor Networks (WSN) have been highly developed which can be used in agriculture to enable optimal irrigation scheduling. Since there is an absence of widely used available methods to support effective agriculture practice in different weather conditions, WSN technology can be used to optimise irrigation in the crop fields. This paper presents architecture of an irrigation system by incorporating interoperable IP based WSN, which uses the protocol stacks and standard of the Internet of Things paradigm. The performance of fundamental issues of this network is emulated in Tmote Sky for 6LoWPAN over IEEE 802.15.4 radio link using the Contiki OS and the Cooja simulator. The simulated results of the performance of the WSN architecture presents the Round Trip Time (RTT) as well as the packet loss of different packet size. In addition, the average power consumption and the radio duty cycle of the sensors are studied. This will facilitate the deployment of a scalable and interoperable multi hop WSN, positioning of border router and to manage power consumption of the sensors.
Wireless Sensor Networks (WSNs) are playing remarkable contribution in real time decision making by actuating the surroundings of environment. As a consequence, the contemporary agriculture is now using WSNs technology for better crop production, such as irrigation scheduling based on moisture level data sensed by the sensors. Since WSNs are deployed in constraints environments, the life time of sensors is very crucial for normal operation of the networks. In this regard routing protocol is a prime factor for the prolonged life time of sensors. This research focuses the performances analysis of some clustering based routing protocols to select the best routing protocol. Four algorithms are considered, namely Low Energy Adaptive Clustering Hierarchy (LEACH), Threshold Sensitive Energy Efficient sensor Network (TEEN), Stable Election Protocol (SEP) and Energy Aware Multi Hop Multi Path (EAMMH). The simulation is carried out in Matlab framework by using the mathematical models of those algortihms in heterogeneous environment. The performance metrics which are considered are stability period, network lifetime, number of dead nodes per round, number of cluster heads (CH) per round, throughput and average residual energy of node. The experimental results illustrate that TEEN provides greater stable region and lifetime than the others while SEP ensures more througput.
Because of the increased popularity and fast expansion of the Internet as well as Internet of things, networks are growing rapidly in every corner of the society. As a result, huge amount of data is travelling across the computer networks that lead to the vulnerability of data integrity, confidentiality and reliability. So, network security is a burning issue to keep the integrity of systems and data. The traditional security guards such as firewalls with access control lists are not anymore enough to secure systems. To address the drawbacks of traditional Intrusion Detection Systems (IDSs), artificial intelligence and machine learning based models open up new opportunity to classify abnormal traffic as anomaly with a self-learning capability. Many supervised learning models have been adopted to detect anomaly from networks traffic. In quest to select a good learning model in terms of precision, recall, area under receiver operating curve, accuracy, F-score and model built time, this paper illustrates the performance comparison between Naïve Bayes, Multilayer Perceptron, J48, Naïve Bayes Tree, and Random Forest classification models. These models are trained and tested on three subsets of features derived from the original benchmark network intrusion detection dataset, NSL-KDD. The three subsets are derived by applying different attributes evaluator’s algorithms. The simulation is carried out by using the WEKA data mining tool.
Glaucoma detection is an important research area in intelligent system and it plays an important role to medical field. Glaucoma can give rise to an irreversible blindness due to lack of proper diagnosis. Doctors need to perform many tests to diagnosis this threatening disease. It requires a lot of time and expense. Sometime affected people may not have any vision loss, at the early stage of glaucoma. For detecting glaucoma, we have built a model to lessen the time and cost. Our work introduces a CNN based Inception V3 model. We used total 6072 images. Among this image 2336 were glaucomatous and 3736 were normal fundus image. For training our model we took 5460 images and for testing we took 612 images. After that we obtained an accuracy of 0.8529 and a value of 0.9387 for AUC. For comparison, we used DenseNet121 and ResNet50 algorithm and got an accuracy of 0.8153 and 0.7761 respectively.
Accurate and rapid identification of the severe and non-severe COVID-19 patients is necessary for reducing the risk of overloading the hospitals, effective hospital resource utilization, and minimizing the mortality rate in the pandemic. A conjunctive belief rule-based clinical decision support system is proposed in this paper to identify critical and non-critical COVID-19 patients in hospitals using only three blood test markers. The experts’ knowledge of COVID-19 is encoded in the form of belief rules in the proposed method. To fine-tune the initial belief rules provided by COVID-19 experts using the real patient’s data, a modified differential evolution algorithm that can solve the constraint optimization problem of the belief rule base is also proposed in this paper. Several experiments are performed using 485 COVID-19 patients’ data to evaluate the effectiveness of the proposed system. Experimental result shows that, after optimization, the conjunctive belief rule-based system achieved the accuracy, sensitivity, and specificity of 0.954, 0.923, and 0.959, respectively, while for disjunctive belief rule base, they are 0.927, 0.769, and 0.948. Moreover, with a 98.85% AUC value, our proposed method shows superior performance than the four traditional machine learning algorithms: LR, SVM, DT, and ANN. All these results validate the effectiveness of our proposed method. The proposed system will help the hospital authorities to identify severe and non-severe COVID-19 patients and adopt optimal treatment plans in pandemic situations.
Emojis are like small icons or images used to express our sentiments or feelings via text messages. They are extensively used in different social media platforms like Facebook, Twitter, Instagram etc. We considered hand-drawn emojis to classify them into 8 classes in this research paper. Hand-drawn emojis are the emojis drawn in any digital platform or in just a paper with a pen. This paper will enable the users to classify the hand-drawn emojis so that they could use them in any social media without any confusion. We made a local dataset of 500 images for each class summing a total of 4000 images of hand-drawn emojis. We presented a system which could recognise and classify the emojis into 8 classes with a convolutional neural network model. The model could favorably recognise as well as classify the hand-drawn emojis with an accuracy of 97%. Some pre-trained CNN models like VGG16, VGG19, ResNet50, MobileNetV2, InceptionV3 and Xception are also trained on the dataset to compare the accuracy and check whether they are better than the proposed one. On the other hand, machine learning models like SVM, Random Forest, Adaboost, Decision Tree and XGboost are also implemented on the dataset.
Mosquitoes are responsible for the most number of deaths every year throughout the world. Bangladesh is also a big sufferer of this problem. Dengue, malaria, chikungunya, zika, yellow fever etc. are caused by dangerous mosquito bites. The main three types of mosquitoes which are found in Bangladesh are aedes, anopheles and culex. Their identification is crucial to take the necessary steps to kill them in an area. Hence, a convolutional neural network (CNN) model is developed so that the mosquitoes could be classified from their images. We prepared a local dataset consisting of 442 images, collected from various sources. An accuracy of 70% has been achieved by running the proposed CNN model on the collected dataset. However, after augmentation of this dataset which becomes 3,600 images, the accuracy increases to 93%. We also showed the comparison of some methods with the CNN method which are VGG-16, Random Forest, XGboost and SVM. Our proposed CNN method outperforms these methods in terms of the classification accuracy of the mosquitoes. Thus, this research forms an example of humanitarian technology, where data science can be used to support mosquito classification, enabling the treatment of various mosquito borne diseases.
Optimization problem like Travelling Salesman Problem (TSP) can be solved by applying Genetic Algorithm (GA) to obtain perfect approximation in time. In addition, TSP is considered as a NP-hard problem as well as an optimal minimization problem. Selection, crossover and mutation are the three main operators of GA. The algorithm is usually employed to find the optimal minimum total distance to visit all the nodes in a TSP. Therefore, the research presents a new crossover operator for TSP, allowing the further minimization of the total distance. The proposed crossover operator consists of two crossover point selection and new offspring creation by performing cost comparison. The computational results as well as the comparison with available well-developed crossover operators are also presented. It has been found that the new crossover operator produces better results than that of other cross-over operators.
The main challenge of any mobile robot is to detect and avoid obstacles and potholes. This paper presents the development and implementation of a novel mobile robot. An Arduino Uno is used as the processing unit of the robot. A Sharp distance measurement sensor and Ultrasonic sensors are used for taking inputs from the environment. The robot trains a neural network based on a feedforward backpropagation algorithm to detect and avoid obstacles and potholes. For that purpose, we have used a truth table. Our experimental results show that our developed system can ideally detect and avoid obstacles and potholes and navigate environments.
An earthquake is a tremor felt on the surface of the earth created by the movement of the major pieces of its outer shell. Till now, many attempts have been made to forecast earthquakes, which saw some success, but these attempted models are specific to a region. In this paper, an earthquake occurrence and location prediction model is proposed. After reviewing the literature, long short-term memory (LSTM) is found to be a good option for building the model because of its memory-keeping ability. Using the Keras tuner, the best model was selected from candidate models, which are composed of combinations of various LSTM architectures and dense layers. This selected model used seismic indicators from the earthquake catalog of Bangladesh as features to predict earthquakes of the following month. Attention mechanism was added to the LSTM architecture to improve the model’s earthquake occurrence prediction accuracy, which was 74.67%. Additionally, a regression model was built using LSTM and dense layers to predict the earthquake epicenter as a distance from a predefined location, which provided a root mean square error of 1.25.
An Internet-of-Things (IoT)-Belief Rule Base (BRB) based hybrid system is introduced to assess Autism spectrum disorder (ASD). This smart system can automatically collect sign and symptom data of various autistic children in realtime and classify the autistic children. The BRB subsystem incorporates knowledge representation parameters such as rule weight, attribute weight and degree of belief. The IoT-BRB system classifies the children having autism based on the sign and symptom collected by the pervasive sensing nodes. The classification results obtained from the proposed IoT-BRB smart system is compared with fuzzy and expert based system. The proposed system outperformed the state-of-the-art fuzzy system and expert system.
Recent advancements in the fields of sensor equipment and wireless sensor networks have opened the window of opportunity for many innovative applications. In this paper, we propose a new architecture for building decision support systems using heterogeneous wireless sensor networks. The architecture is built around standard hardware and existing wireless sensor networks technology. We show the effectiveness of the proposed architecture by applying it to a flood prediction scenario.
Smart risk assessment systems are becoming more and more important in the society. If the chances of reducing and managing certain risks are increased, the impacts can be controlled and reduced significantly. This article surveys different belief-rule-based decision support systems and various wireless sensor network technologies that can be used in collaboration to build interesting risk assessment applications. We propose a model for building such an environment and describe a potential application of our proposed model for assessing flood risks in a case study.
Capability approach is hard to operationalize and capabilities are sometimes difficult to reduce into the individual level. We describe our approach to solve these problems paired with an adequate ICT infrastructure and believe that the impact of concrete eGovernment services will become more visible and available. Citizens can inspect it, professionals can target their re-engineering on failure cases, learn from best cases, and researchers can compare different types of eGovernment. The solving of operationalisation will also strengthen the capability approach, by extending its scope from macro theory towards more easy application on the institutional and organizational levels.
Skin cancer has been one of the top three cancers that can be fatal when caused by broken DNA. Damaged DNA causes cells to expand uncontrollably, and the rate of growth is currently increasing rapidly. Some studies have been conducted on the computerized detection of malignancy in skin lesion images. However, due to some problematic aspects such as light reflections from the skin surface, differences in color lighting, and varying forms and sizes of the lesions, analyzing these images is extremely difficult. As a result, evidence-based automatic skin cancer detection can help pathologists improve their accuracy and competency in the early stages of the disease. In this paper, we present a transfer ring strategy based on a convolutional neural network (CNN) model for accurately classifying various types of skin lesions. Preprocessing normalizes the input photos for accurate classification; data augmentation increases the amount of images, which enhances classification rate accuracy. The performance of the GoogLeNet transfer learning model is compared to that of other transfer learning models such as Xpection, InceptionResNetVe, and DenseNet, among others. The model was tested on the ISIC dataset, and we ended up with the highest training and testing accuracy of 91.16% and 89.93%, respectively. When compared to existing transfer learning models, the final results of our proposed GoogLeNet transfer learning model characterize it as more dependable and resilient.
Non small cell Lung cancer (NSCLC) is one of the most well-known types of Lung cancer which is reason for cancer related demise in Bangladesh. The early detection stage of NSCLC is required for improving the survival rate by taking proper decision for surgery and radiotherapy. The most common factors for staging NSCLC are age, tumor size, lymph node distance, Metastasis and Co morbidity. Moreover, physicians' diagnosis is unable to give more reliable outcome due to some uncertainty such as ignorance, incompleteness, vagueness, randomness, imprecision. Belief Rule Base Expert System (BRBES) is fit to deal with above mentioned uncertainty by applying both Belief Rule base and Evidential Reasoning approach. Therefore, this paper represents the architecture, development and interface for staging NSCLC by incorporating belief rule base as well as evidential reasoning with the capability of handling uncertainty. At last, a comparative analysis is added which indicate that the outcomes of proposed expert system is more reliable and efficient than the outcomes generated from traditional human expert as well as Support Vector Machine (SVM) or Fuzzy Rule Base Expert System (FRBES).
This paper proposes a process of sentiment analysis of movie reviews written in Bangla language. This process can automate the analysis of audience’s reaction towards a specific movie or TV show. With more and more people expressing their opinions openly in the social networking sites, analyzing the sentiment of comments made about a specific movie can indicate how well the movie is being accepted by the general public. The dataset used in this experiment was collected and labeled manually from publicly available comments and posts from social media websites. Using Support Vector Machine algorithm, this model achieves 88.90% accuracy on the test set and by using Long Short Term Memory network [1] the model manages to achieve 82.42% accuracy. Furthermore, a comparison with some other machine learning approaches is presented in this paper.
This paper proposes a process of Handwritten Character Recognition to recognize and convert images of individual Bangla handwritten characters into electronically editable format, which will create opportunities for further research and can also have various practical applications. The dataset used in this experiment is the BanglaLekha-Isolated dataset [1]. Using Convolutional Neural Network, this model achieves 91.81% accuracy on the alphabets (50 character classes) on the base dataset, and after expanding the number of images to 200,000 using data augmentation, the accuracy achieved on the test set is 95.25%. The model was hosted on a web server for the ease of testing and interaction with the model. Furthermore, a comparison with other machine learning approaches is presented.
Current water quality monitoring system is a manual system with a monotonous process and is very time-consuming. This paper proposes a sensor-based water quality monitoring system. The main components of Wireless Sensor Network (WSN) include a microcontroller for processing the system, communication system for inter and intra node communication and several sensors. Real-time data access can be done by using remote monitoring and Internet of Things (IoT) technology. Data collected at the apart site can be displayed in a visual format on a server PC with the help of Spark streaming analysis through Spark MLlib, Deep learning neural network models, Belief Rule Based (BRB) system and is also compared with standard values. If the acquired value is above the threshold value automated warning SMS alert will be sent to the agent. The uniqueness of our proposed paper is to obtain the water monitoring system with high frequency, high mobility, and low powered. Therefore, our proposed system will immensely help Bangladeshi populations to become conscious against contaminated water as well as to stop polluting the water.
Investors in the stock market have always been in search of novel and unique techniques so that they can successfully predict stock price movement and make a big profit. However, investors continue to look for improved and new techniques to beat the market instead of old and traditional ones. Therefore, researchers are continuously working to build novel techniques to supply the demand of investors. Different types of recurrent neural networks (RNN) are used in time series analyses, especially in stock price prediction. However, since not all stocks’ prices follow the same trend, a single model cannot be used to predict the movement of all types of stock’s price. Therefore, in this research we conducted a comparative analysis of three commonly used RNNs—simple RNN, Long Short Term Memory (LSTM), and Gated Recurrent Unit (GRU)—and analyzed their efficiency for stocks having different stock trends and various price ranges and for different time frequencies. We considered three companies’ datasets from 30 June 2000 to 21 July 2020. The stocks follow different trends of price movements, with price ranges of $30, $50, and $290 during this period. We also analyzed the performance for one-day, three-day, and five-day time intervals. We compared the performance of RNN, LSTM, and GRU in terms of R2 value, MAE, MAPE, and RMSE metrics. The results show that simple RNN is outperformed by LSTM and GRU because RNN is susceptible to vanishing gradient problems, while the other two models are not. Moreover, GRU produces lesser errors comparing to LSTM. It is also evident from the results that as the time intervals get smaller, the models produce lower errors and higher reliability.
Surviving three years of the pandemic since December 2019, monitoring COVID-19 patients in a projected way is still challenging. Even after testing negative for coronavirus, people face a lot of post-covid stresses and symptoms. Scarcity of hospital beds, shortage of medical equipment like oxygen, ventilation, etc. have made the situation worse as people failed to receive proper treatment. In this regard, this work proposes an IoMT-based wearable checking device for assessing COVID-19-identified imperative signals. Furthermore, by continuously monitoring data, the device promptly warns concerned clinical personnel about any breach of isolation for possibly contaminated patients. The data from the body-wearable sensor is processed and broken down by an edge node in the IoMT cloud to characterize the condition of health. A puttable IoMT sensor layer, a cloud layer with Application Peripheral Interface (API), and an Android-based cell prototype are part of the proposed system. Each layer has its own function; for example, the data from the IoMT sensor layer is used to characterize the wellness of the side effects. The Android portable application layer is in charge of informing and cautioning possibly infected patient family members, the nearest hospital, and the patient’s signed doctor about the potential contamination. Two APIs and a variety of applications are synchronized in the integrated system to predict and disrupt the situation. In a word, the target is to monitor this data and send it to the cloud through the IoMT gateway and monitor these parameters using the Android app. The doctor and the patient’s relative could also observe the monitor system through the app using the device id from this app. Because there are fewer available beds in hospitals, more people are dying as a result of inadequate care.
Recommendation system can predict the ratings of users to items by leveraging machine learning algorithms. The use of recommendation systems is common in e-commerce websites now-a-days. Since enormous amounts of data including users’ click streams, purchase history, demographics, social networking comments and user-item ratings are stored in e-commerce systems databases, the volume of the data is getting bigger at high speed, and the data is sparse. However, the recommendations and predictions must be made in real time, enabling to bring enormous benefits to human beings. Apache spark is well suited for applications which require high speed query of data, transformation and analytics results. Therefore, the recommendation system developed in this research is implemented on Apache Spark. Also, the matrix factorization using Alternating Least Squares (ALS) algorithm which is a type of collaborative filtering is used to solve overfitting issues in sparse data and increases prediction accuracy. The overfitting problem arises in the data as the user-item rating matrix is sparse. In this research a recommendation system for e-commerce using alternating least squares (ALS) matrix factorization method on Apache Spark MLlib is developed. The research shows that the RMSE value is significantly reduced using ALS matrix factorization method and the RMSE is 0.870. Consequently, it is shown that the ALS algorithm is suitable for training explicit feedback data set where users provide ratings for items.
Though speech recognition has been a common interest of researchers over the last couple of decades, but very few works have been done on Bangla voice recognition. In this research, we developed a digital personal assistant for handicapped people which recognizes continuous Bangla voice commands. We employed the cross-correlation technique which compares the energy of Bangla voice commands with prerecorded reference signals. After recognizing a Bangla command, it executes a task specified by that command. Mouse cursor can also be controlled using the facial movement of a user. We validated our model in three different environments (noisy, moderate and noiseless) so that the model can act naturally. We also compared our proposed model with a combined model of MFCC & DTW, and another model which combines crosscorrelation with LPC. Results indicate that the proposed model achieves a huge accuracy and smaller response time comparing to the other two techniques.
Researchers around the world are now focused on to make our devices more interactive and trying to make the devices operational with minimal physical contact. In this research, we propose an interactive computer system which can operate without any physical keyboard and mouse. This system can be beneficial to everyone, especially to the paralyzed people who face difficulties to operate physical keyboard and mouse. We used computer vision so that user can type on virtual keyboard using a yellow-colored cap on his fingertip, and can also navigate to mouse controlling system. Once the user is in mouse controlling mode, user can perform all the mouse operations only by showing different number of fingers. We validated both module of our system by a 52 years old paralyzed person and achieved around 80% accuracy on average.
Price prediction of financial assets has been a key interest for researchers over the decades. Numerous techniques to predict the price movements have been developed by the researchers over the years. But a model loses its credibility once a large number of traders start using the same technique. Therefore, the traders are in continuous search of new and efficient prediction techniques. In this research, we propose a novel machine learning technique using technical analysis with Belief Rule-Based Expert System (BRBES), and incorporating the concept of Bollinger Band to forecast stock price in the next five days. A Bollinger Event is triggered when the closing price of the stock goes down the Lower Bollinger Band. The BRBES approach has never been applied to stock markets, despite its potential and the appetite of the financial markets for expert systems. We predict the price movement of the Swedish company TELIA as a proof of concept. The knowledge base of the initial BRBES is constructed by simulating the historical data and then the learning parameters are optimized using MATLAB’s fmincon function. We evaluate the performance of the trained BRBES in terms of Accuracy, Area Under ROC Curve, Root Mean Squared Error, type I error, type II error, value, and profit/loss ratio. We compare our proposed model against a similar rule-based technique, Adaptive Neuro-Fuzzy Inference System (ANFIS), to understand the significance of the improved rule base of BRBES. We also compare the performance against Support Vector Machine (SVM), one of the most popular machine learning techniques, and a simple heuristic model. Finally, the trained BRBES is compared against recent state-of-the-art deep learning approaches to show how competitive the performance of our proposed model is. The results show that the trained BRBES produces better performance than the non-trained BRBES, ANFIS, SVM, and the heuristic approaches. Also, it indicates better or competitive performance against the deep learning approaches. Thus BRBES exhibits its potential in predicting financial asset price movement.
In accordance with the World Health Organization’s instruction, the air quality in Bangladesh is considered perilous. A productive and precise air quality index (AQI) is a must and one of the obligatory conditions for helping the society to be viable in lieu of the consequences of air contamination. If we know the index of air quality in advance, then it would be of a great help saving our health from air contamination. This study introduces an air quality index prediction model for two mostly polluted cities in Bangladesh: Dhaka and Chattogram. Gated recurrent unit (GRU), long short-term memory (LSTM) are the two robust variation of recurrent neural network (RNN). This model combines these two together. We have used GRU as first hidden layer and LSTM as the second hidden layer of the model, followed by two dense layers. After collecting and processing the data, the model was trained on 80% of the data and then validated against the remaining data. We have evaluated the performance of the model considering MSE, RMSE, and MAE to see how much error does the model produce. Results reflect that our model can follow the actual AQI trends for both cities. At last, we have juxtaposed the performance of our proposed hybrid model against a standalone GRU model and a standalone LSTM model. Results also show that combining these two models improves the overall model’s performance.
Human may have multiple reactions at a time. Social media is a real-life example where people can express their reactions or opinions. For example, Facebook has become a widely used social media where users express their opinion on different posts such as status or photo post. Therefore, user opinions can easily be achieved through social media and then analyzed and applied in different practical fields. Reaction assessment of Social media can be an excellent source of information. Therefore, an accurate assessment of human reaction is a must. Facebook provides six emoticons for each of the posts of its users. The six emoticons define six types of reactions. The user, who wants to react to any post, have to choose only one of the six emoticons. There is no scope for a user to express multiple reactions at a time. Moreover, if a user selects an emoticon, then the system takes that input as 100% of that corresponding reaction, or 0% if not selected, thus the reaction assessment system becomes a Boolean system. Therefore, the assessment of multiple reactions of the user cannot be measured with 100% certainty due to the existence of various types of uncertainties such as vagueness, imprecision, randomness, ignorance, incompleteness, and ambiguity in the system. Therefore, to assess multiple human reactions, an expert system is needed to handle all these uncertainties. The system design, development process, and applications of an expert system to assess multiple human reactions are described in this paper. For the development of the expert system, the Belief Rule-Based Inference Methodology using the Evidential Reasoning (RIMER) approach has been used and the system is named as a Belief Rule-Based Expert System (BRBES). The developed BRBES can mitigate all the uncertainties mentioned above.
The primary diagnosis of Tuberculosis (TB) is usually carried out by looking at the various signs and symptoms of a patient. However, these signs and symptoms cannot be measured with 100\% certainty since they are associated with various types of uncertainties such as vagueness, imprecision, randomness, ignorance and incompleteness. Consequently, traditional primary diagnosis, based on these signs and symptoms, which is carried out by the physicians, cannot deliver reliable results. Therefore, this article presents the design, development and applications of a Belief Rule Based Expert System (BRBES) with the ability to handle various types of uncertainties to diagnose TB. The knowledge base of this system is constructed by taking experts' suggestions and by analyzing historical data of TB patients. The experiments, carried out, by taking the data of 100 patients demonstrate that the BRBES's generated results are more reliable than that of human expert as well as fuzzy rule based expert system.
Schizophrenia is a severe neurological disease where a patient’s perceptions of reality are disrupted. Its symptoms include hallucinations, delusions, and profoundly strange thinking and behavior, which make the patient’s daily functions difficult. Despite identifying genetic variations linked to Schizophrenia, causative genes involved in pathogenesis and expression regulations remain unknown. There is no particular way in life sciences for diagnosing Schizophrenia. Commonly used machine learning and deep learning are data-oriented. They lack the ability to deal with uncertainty in data. Belief Rule Based Expert System (BRBES) methodology addresses various categories of uncertainty in data with evidential reasoning. Previous researches showed the association of DNA methylation (DNAm) with risk of Schizophrenia. Whole blood DNAm data, hence, is useful for smart diagnosis of Scizophrenia. However, to our knowledge, no previous studies have investigated the performance of BRBES to diagnose Schizophrenia. Therefore, in this study, we explore BRBES’ performance in diagnosing Schizophrenia using whole blood DNAm data. BRBES was optimized by gradient-free algorithms due to the limitations of gradient-based optimization. Classification thresholds were optimized to yield better results. Finally, we compared performance to two machine learning models after 5-fold cross-validation where our model achieved the highest average sensitivity (76.8%) among the three.
The impact of earthquake is devastating, which has the capability to stop the socio-economic activities of a region within a short span of time. Therefore, an earlier prediction of earthquake could play an important role to save human lives as well as socio-economic activities. The signs of animal behavior along with environmental and chemical changes in nature could be considered as a way to predict the earthquake. These factors cannot be determined accurately because of the presence of different categories of uncertainties. Therefore, this article presents a belief rule based expert system (BRBES) which has the capability to predict earthquake under uncertainty. Historical data of various earthquakes of the world with specific reference to animal behavior as well as environmental and chemical changes have been considered in validating the BRBES. The reliability of our proposed BRBES’s output is measured in comparison with Fuzzy Logic Based Expert System (FLBES) and Artificial Neural Networks (ANN) based system, whereas our BRBES’s results are found more reliable than that of FLBES and ANN. Therefore, this BRBES can be considered to predict the occurrence of an earthquake in a region by taking account of the data, related to the animal, environmental and chemical changes.
Measles is a highly infectious child disease that causes serious complications and death worldwide. Measles is generally diagnosed from its signs and symptoms by a physician, which cannot be measured with 100% certainty during the diagnosis process. Consequently, the traditional way of diagnosing measles from its signs and symptoms lacks the accuracy. Therefore, a belief rule-based inference methodology using evidential reasoning approach (RIMER), which is capable of handling various types of uncertainties has been used to develop an expert system to diagnose measles under uncertainty. The results, generated, from the system have been compared with the expert opinion as well as with a Fuzzy Logic based system. In both the cases, it has been found that the Belief Rule Based Expert (BRBES), presented in this paper, is more reliable and accurate.
Dengue Fever is a debilitating mosquito-borne disease, causing sudden fever, leading to fatality in many cases. A Dengue patient is diagnosed by the physicians by looking at the various signs, symptoms and risk factors of this disease. However, these signs, symptoms and the risk factors cannot be measured with 100% certainty since various types of uncertainties such as imprecision, vagueness, ambiguity, and ignorance are associated with them. Hence, it is difficult for the physicians to diagnose the dengue patient accurately since they don’t consider the uncertainties as mentioned. Therefore, this paper presents the design, development and applications of an expert system by incorporating belief rule base as the knowledge representation schema as well as the evidential reasoning as the inference mechanism with the capability of handling various types of uncertainties to diagnose dengue fever. The results generated from the expert system are more reliable than from fuzzy rule based system or from human expert.
A rapidly emerging trend in the IT landscape is the uptake of large-scale datacenters moving storage and data processing to providers located far away from the end-users or locally deployed servers. For these large-scale datacenters, power efficiency is a key metric, with the PUE (Power Usage Effectiveness) and DCiE (Data Centre infrastructure Efficiency) being important examples. This article proposes a belief rule based expert system to predict datacenter PUE under uncertainty. The system has been evaluated using real-world data from a data center in the UK. The results would help planning construction of new datacenters and the redesign of existing datacenters making them more power efficient leading to a more sustainable computing environment. In addition, an optimal learning model for the BRBES demonstrated which has been compared with ANN and Genetic Algorithm; and the results are promising.
Acute coronary syndrome (ACS) is responsible for the obstruction of coronary arteries, resulting in the loss of lives. The onset of ACS can be determined by looking at the various signs and symptoms of a patient. However, the accuracy of ACS determination is often put into question since there exist different types of uncertainties with the signs and symptoms. Belief rule-based expert systems (BRBESs) are widely used to capture uncertain knowledge and to accomplish the task of reasoning under uncertainty by employing belief rule base and evidential reasoning. This article presents the process of developing a BRBES to determine ACS predictability. The BRBES has been validated against the data of 250 patients suffering from chest pain. It is noticed that the outputs created from the BRBES are more dependable than that of the opinion of cardiologists as well as other two expert system tools, namely artificial neural networks and support vector machine. Hence, it can be argued that the BRBES is capable of playing an important role in decision making as well as in avoiding costly laboratory investigations. A procedure to train the system, allowing its enhancement of performance, is also presented.
Chikungunya is a virus-related disease, bring about by the virus called CHIKV that spreads throughmosquito biting. This virus first found in Tanzania, while blood from patients was isolated. Thecommon signs and symptoms, associated with Chikungunya are considered as fever, joint swelling,joint pain, muscle pain and headache. The examination of these signs and symptoms by the physician constitutes the typical preliminary diagnosis of this disease. However, the physician is unable tomeasure them with accuracy. Therefore, the preliminary diagnosis in most of the cases could sufferfrom inaccuracy, which leads to wrong treatment. Hence, this paper introduces the design and implementation of a belief rule based expert system (BRBES) which is capable to represent uncertainknowledge as well as inference under uncertainty. Here, the knowledge is illustrated by employing belief rule base while deduction is carried out by evidential reasoning. The real patient data of250 have been considered to demonstrate the accuracy and the robustness of the expert system. Acomparison has been performed with the results of BRBES and Fuzzy Logic Based Expert System(FLBES) as well as with the expert judgment. Furthermore, the result of BRBES has been contrastedwith various data-driven machine learning approaches, including ANN (Artificial Neural networks)and SVM (Support Vector Machine). The reliability of BRBESs was found better than those of datadriven machine learning approaches. Therefore, the BRBES presented in this paper could enable thephysician to conduct the analysis of Chikungunya more accurately.
Hypertension (HPT) plays an important role, especially for stroke and heart diseases. Therefore, theaccurate assessment of hypertension is becoming a challenge. However, the presence of uncertainties, associated with the signs and symptoms of HPT are becoming crucial to conduct the preciseassessment. This article presents a web-based expert system (web BRBES) by employing beliefrule based (BRB) methodology to assess HPT, allowing the generation of reliable results. In order tocheck the reliability of the system, a comparison has been performed among various approaches suchas decision tree, random forest, artificial neural networks, fuzzy rule based expert system and experts’opinion. Different performance metrics such as confusion matrix, accuracy, root mean square error,area under curve have been used to contrast the reliability of the approaches. The BRBES producesa more reliable result than from the other approaches. Moreover, the user friendliness of the webBRBES found high as obtained by using the PACT (People, Activities, Contexts, Technologies) approach over 200 people.
Natural calamities such as flooding, volcanic eruption, tornado hampers our daily life and causes many sufferings. Flood is one of the most catastrophic among the natural calamities. Assessing flood risk helps us to take necessary steps and save human lives. Several heterogeneous factors are used to assess flood risk on the livelihood of an area. Moreover, several types of uncertainties can be associated with each factor. In this paper, we propose a web based flood risk assessment expert system by combining belief rule base with the capability of reading data and generating web-based output. This paper also introduces a generic RESTful API which can be used without writing the belief rule based expert system from scratch. This expert system will facilitate the monitoring of the various flood risk factors, contributing in increasing the flood risk on livelihood of an area. Eventually, the decision makers should be able to take measures to control those factors and to reduce the risk of flooding in an area. Data for the expert system has been collected from a case study area by conducting interviews.
The Covid-19 disease which was caused by novel coronavirus (SARS-CoV-2) has already become a great threat for humans beings. The virus is spreading rapidly around the world. Therefore, we crucially need quick diagnostic tests to identify affected patients and to minimize the spread of the virus. With the advancements of Machine Learning, the detection of Covid-19 in the early stage would facilitate taking precautions as early as possible. However, because of the lack of data-sets, especially chest X-ray images of Covid-19 affected patients, it has become challenging to detect this disease. In this paper, a deep transfer learning-based pre-trained model is named VGG16 along with adapt histogram equalization has been developed to diagnose Covid-19 by using X-ray images. An image processing technique named adaptive histogram equalization has been used to generate more images by using the existing data set. It can be observed that VGG-16 provides the highest accuracy which is 98.75% in comparison to two other pre-trained models such as VGG-19 and Mobilnenet-V2(97% accuracy for VGG-19, 92.65% accuracy for Mobilenet-V2).
In recent years, the foreign exchange (FOREX) market has attracted quite a lot of scrutiny from researchers all over the world. Due to its vulnerable characteristics, different types of research have been conducted to accomplish the task of predicting future FOREX currency prices accurately. In this research, we present a comprehensive review of the recent advancements of FOREX currency prediction approaches. Besides, we provide some information about the FOREX market and cryptocurrency market. We wanted to analyze the most recent works in this field and therefore considered only those papers which were published from 2017 to 2019. We used a keyword-based searching technique to filter out popular and relevant research. Moreover, we have applied a selection algorithm to determine which papers to include in this review. Based on our selection criteria, we have reviewed 39 research articles that were published on “Elsevier”, “Springer”, and “IEEE Xplore” that predicted future FOREX prices within the stipulated time. Our research shows that in recent years, researchers have been interested mostly in neural networks models, pattern-based approaches, and optimization techniques. Our review also shows that many deep learning algorithms, such as gated recurrent unit (GRU) and long short term memory (LSTM), have been fully explored and show huge potential in time series prediction.
Computer is a part and parcel in our day to day life and used in various fields. The interaction of human and computer is accomplished by traditional input devices like mouse, keyboard etc. Hand gestures can be a useful medium of human-computer interaction and can make the interaction easier. Gestures vary in orientation and shape from person to person. So, non-linearity exists in this problem. Recent research has proved the supremacy of Convolutional Neural Network (CNN) for image representation and classification. Since, CNN can learn complex and non-linear relationships among images, in this paper, a static hand gesture recognition method using CNN was proposed. Data augmentation like re-scaling, zooming, shearing, rotation, width and height shifting was applied to the dataset. The model was trained on 8000 images and tested on 1600 images which were divided into 10 classes. The model with augmented data achieved accuracy 97.12% which is nearly 4% higher than the model without augmentation (92.87%).
Natural calamity disrupts our daily life and brings many sufferings in our life. Among the natural calamities, flood is one of the most catastrophic. Predicting flood helps us to take necessary precautions and save human lives. Several types of data (meteorological condition, topography, river characteristics, and human activities) are used to predict flood water level in an area. In our previous works, we proposed a belief rule based flood prediction system in a desktop environment. In this paper, we propose a web-service based flood prediction expert system by incorporating belief rule base with the capability of reading sensor data such as rainfall, river flow on real time basis. This will facilitate the monitoring of the various flood-intensifying factors, contributing in increasing the flood water level in an area. Eventually, the decision makers would able to take measures to control those factors and to reduce the intensity of flooding in an area.
5G networks will enable people and machines to communicate at high speeds and very low latencies, in a reliable way. This opens up opportunities for totally new usage patterns and the fully connected Industry 4.0-enabled enterprise covering the entire value chain from design, production, deployment, to usage of products. 5G will be rolled out across the whole world, including Sweden where the first 5G test network was launched late 2018. One important new feature in 5G is the emerging edge computing capabilities, where users can easily offload computational tasks to the network’s edge very close to the user. At the same time, computational tasks traditionally performed in central nodes can be offloaded from remotely located data centres to the network’s edge. Multi-access Edge Computing (MEC) is a promising network architecture delivering solutions along these lines offering a platform for applications with requirements on low latencies and high reliability. This paper targets this environment with a novel Belief-rule-based (BRB) unsupervised learning algorithm for clustering helping 5G applications to take intelligent decisions on software deployment. The scenarios consist of different combinations of numbers of users and connections and mobility patterns. The target environment is built up using a three-tier structure with a container-based solution where software components can easily be spread around the network.
Recent technological advancements in the area of the Internet of Things (IoT) and cloud services, enable the generation of large amounts of raw data. However, the accurate prediction by using this data is considered as challenging for machine learning methods. Deep Learning (DL) methods are widely used to process large amounts of data because they need less preprocessing than traditional machine learning methods. Various types of uncertainty associated with large amounts of raw data hinder the prediction accuracy. Belief Rule-Based Expert Systems (BRBES) are widely used to handle uncertain data. However, due to their incapability of integrating associative memory within the inference procedures, they demonstrate poor accuracy of prediction when large amounts of data is considered. Therefore, we propose the integration of an associative memory based DL method within the BRBES inference procedures, allowing to discover accurate data patterns and hence, the improvement of prediction under uncertainty. To demonstrate the applicability of the proposed method, which is named BRB-DL, it has been fine tuned against two datasets, one in the area of air pollution and the other in the area of power generation. The reliability of the proposed BRB-DL method, has also been compared with other DL methods such as Long-Short Term Memory and Deep Neural Network, and BRBES by taking into account of the air quality dataset from Beijing city and the power generation dataset of a combined cycle power plant. BRB-DL outperforms the above-mentioned methods in terms of prediction accuracy. For example, the Mean Square Error value of BRB-DL is 4.12 whereas for Long-Short Term Memory, Deep Neural Network, Fuzzy Deep Neural Network, Adaptive Neuro Fuzzy Inference System and BRBES it is 18.66, 28.49, 17.05, 16.37 and 38.15 for combined cycle power plant respectively, which are significantly higher.
Nowadays, belief rule-based expert systems (BRBESs) are widely used in various domains which provides a framework to handle qualitative and quantitative data by addressing several kinds of uncertainty. Learning plays an important role in BRBES to upgrade its knowledge base and parameters values, necessary for the improvement of the prediction accuracy. Different optimal training procedures such as Particle Swarm Optimisation (PSO), Differential Evolution (DE), and Genetic Algorithm (GA) have been used as learning mechanisms. Among these procedures, DE performs comparatively better than others. However, DE's performance depends significantly in assigning near optimal values to its control parameters including cross over and mutation factors. Therefore, the objective of this article is to present a novel optimal training procedure by integrating DE with BRBES. This is named as enhanced belief rule-based adaptive differential evolution (eBRBaDE) algorithm because it has the ability to determine the near-optimal values of both the control parameters while ensuring the balanced exploitation and exploration in the search space. In addition, a new joint optimization learning mechanism by using eBRBaDE is presented where both parameter and structure of BRBES are considered. The reliability of the eBRBaDE has been compared with evolutionary optimization algorithms such as GA, PSO, BAT, DE and L-SHADE. This comparison has been carried out by taking account of both conjunctive and disjunctive BRBESs while predicting the Power Usage Effectiveness (PUE) of a datacentre. The comparison demonstrates that the eBRBaDE provides higher prediction accuracy of PUE than from other evolutionary optimization algorithms. Contribution-An enhanced differential evolution algorithm has been proposed in this paper, which is later used as a novel optimal training procedure for BRBES.
It is an era of Internet of Things, where various types of sensors, especially wireless, are widely used to collect huge amount of data to feed various systems such as surveillance, environmental monitoring, and disaster management. In these systems, wireless sensors are deployed to make decisions or to predict an event in a real-time basis. However, the accuracy of such decisions or predictions depends upon the reliability of the sensor data. Unfortunately, erroneous data are received from the sensors. Consequently, it hampers the appropriate operations of the mentioned systems, especially in making decisions and prediction. Therefore, the detection of anomaly that exists with the sensor data drew significant attention and hence, it needs to be filtered before feeding a system to increase its reliability in making decisions or prediction. There exists various sensor anomaly detection algorithms, but few of them are able to address the uncertain phenomenon, associated with the sensor data. If these uncertain phenomena cannot be addressed by the algorithms, the filtered data into the system will not be able to increase the reliability of the decision-making process. These uncertainties may be due to the incompleteness, ignorance, vagueness, imprecision and ambiguity. Therefore, in this paper we propose a new belief-rule-based association rule (BRBAR) with the ability to handle the various types of uncertainties as mentioned.The reliability of this novel algorithm has been compared with other existing anomaly detection algorithms such as Gaussian, binary association rule and fuzzy association rule by using sensor data from various domains such as rainfall, temperature and cancer cell data. Receiver operating characteristic curves are used for comparing the performance of our proposed BRBAR with the aforementioned algorithms. The comparisons demonstrate that BRBAR is more accurate and reliable in detecting anomalies from sensor data under uncertainty. Hence, the use of such algorithm to feed the decision-making systems could be beneficial. Therefore, we have used this algorithm to feed appropriate sensor data to our recently developed belief-rule-based expert system to predict flooding in an area. Consequently, the reliability and the accuracy of the flood prediction system increase significantly. Such novel algorithm (BRBAR) can be used in other areas of applications.
Floods are one of the most dangerous catastrophic events. By the year 2050 flooding due to rise of ocean level may cost one trillion USD to coastal cities. Since flooding involves multi-dimensional elements, its accurate prediction is difficult. In addition, the elements cannot be measured with 100% accuracy. Belief rule-based expert systems (BRBESs) can be considered as an appropriate approach to handle this type of problem because they are capable of addressing uncertainty. However, BRBESs need to be equipped with the capacity to handle multi- level learning and inference to improve its accuracy of flood prediction. Therefore, this paper proposes a new learning and inference mechanism, named joint optimization using belief rule- based adaptive differential evolution (BRBaDE) for multi-level BRBES, which has the capability to handle multi-level learning and inference. Various machine learning methods, including Artificial Neural Networks (ANN), Support Vector Machine (SVM), Linear Regression and Long Short Term Memory have been compared with BRBaDE. The result exhibits that our proposed learning mechanism performs betters than learning techniques as mentioned above in terms of accuracy in flood prediction.
Big Data applications have become increasingly popular with the emergence of cloud computing and the explosion of artificial intelligence. The increasing adoption of data-intensive machines and services is driving the need for more power to keep the data centers of the world running. It has become crucial for large IT companies to monitor the energy efficiency of their data-center facilities and to take actions on the optimization of these heavy electricity consumers. This paper proposes a Belief Rule-Based Expert System (BRBES)-based predictive model to predict the Power Usage Effectiveness (PUE) of a data center. The uniqueness of this model consists of the integration of a novel learning mechanism consisting of parameter and structure optimization by using BRBES-based adaptive Differential Evolution (BRBaDE), significantly improving the accuracy of PUE prediction. This model has been evaluated by using real-world data collected from a Facebook data center located in Luleå, Sweden. In addition, to prove the robustness of the predictive model, it has been compared with other machine learning techniques, such as an Artificial Neural Network (ANN) and an Adaptive Neuro Fuzzy Inference System (ANFIS), where it showed a better result. Further, due to the flexibility of the BRBES-based predictive model, it can be used to capture the nonlinear dependencies of many variables of a data center, allowing the prediction of PUE with much accuracy. Consequently, this plays an important role to make data centers more energy-efficient.
Technological innovation capability (TIC) is a complicated and subtle concept which is based on multiple quantitative and qualitative criteria. The cores of a firm’s long-term competitive dominance are defined by technological innovation capability which is the incentive for a firm’s innovation. Various types of uncertainty can be noticed while considering multiple criteria for evaluating TIC. In order to evaluate TIC in a reliable way, a Belief Rule Base (BRB) Expert System can be used to handle both quantitative and qualitative data and their associated uncertainties. In this paper, a RESTful API-based BRB expert system is introduced to evaluate technological innovation capability by taking uncertainties into consideration. This expert system will facilitate firms’ managers to obtain a recapitulation of the TIC evaluation. It will help them to take essential steps to ensure corporate survival and strengthen their weak capabilities continuously to facilitate a competitive advantage. Other users can also use this API to apply BRB for a different domain. However, a comparison between the knowledge-driven approach (BRBES) and several data-driven models has been performed to find out the reliability in evaluating TIC. The result shows that the outcome of BRBES is better than other data-driven approaches.
Particle Swarm Optimization (PSO) is a meta-heuristic algorithm which is successfully applied to enormous applications to solve non-linear and complex high-dimensional optimization problems. The performance of PSO is greatly influenced by the tuning parameters. Due to the presence of uncertainty or noise in the optimization problem of different domains and to maintain a balance between exploration and exploitation in the search space, these parameters need to be adjusted. Therefore, it is crucial to identify the optimal values of the tuning parameters. In this paper, a new Belief Rule-Based Adaptive Particle Swarm Optimization (BRBAPSO) is proposed where the tuning parameters are adjusted dynamically by considering uncertainties, which ensure a balance between exploitation and exploration in the search space. Two Variants of BRBAPSO, namely Conjunctive BRBAPSO and Disjunctive BRBAPSO, are introduced and they are compared with Time-Varying Inertia Weight PSO (TVIW-PSO), Time-Varying Acceleration Coefficient PSO (TVAC-PSO), and Fuzzy Adaptive PSO (FAPSO) using the CEC 2013 real-parameter optimization benchmark functions. The results show that both variants of BRBAPSO outperform other algorithms on the benchmark functions.