Search

Abstract Myocardial infarction (MI) is a life-threatening medical condition that necessitates both timely and precise diagnosis. The enhancement of automated method to detect MI diseases from Normal patients can play a crucial role in healthcare. This paper presents a novel approach that utilizes the Discrete Wavelet Transform (DWT) for the detection of myocardial signals. The DWT is employed to break down ECG signals into distinct frequency components and subsequently to selectively filter out noise by thresholding the high-frequency details, resulting in denoised ECG signals for myocardial signal detection. These denoised signals are fed into lightweight one-dimensional Convolutional Neural Networks (CNN) for binary classification into Myocardial Infarction (MI) and Normal categories. The paper explores three distinct approaches: utilizing all signals, incorporating under-sampling and up-sampling to address class imbalances, with both noised and denoised signals. Evaluation of the suggested model is done with the help of two publicly available datasets: PTB-XL, a large publicly available electrocardiography dataset and PTB Diagnostic ECG Database. Results obtained through 5-fold cross-validation on the trained model show that the model has achieved an accuracy of 96%, precision of 97% and F1 score of 95%. On cross-validation with the PTB-ECG dataset, this paper achieved an accuracy of 91.18%.

This research addresses the critical issue of women’s safety in urban environments, emphasizing the need for innovative solutions to establish secure pathways. SafeRoutes presents a holistic approach, integrating advanced clustering methodologies and GPS technology, detailing its relevance, ideation, methodology, and anticipated results. During ideation, the team prioritized integrating cutting-edge technologies—artificial intelligence, data analytics, and cloud computing. Emphasizing the constraints of existing safety solutions, the focus was on crafting a sophisticated framework for detailed assessments and real-time risk detection during transit. SafeRoutes aims to redefine women’s safety, providing actionable insights for urban planning and law enforcement. The methodology comprises three integral components. Firstly, a robust data ingestion pipeline connects to public and government data sources, ensuring near real-time models enriched with the latest data. The second component uses unsupervised machine learning models, comparing and employing various clustering algorithms. Parameters like crime rates, police presence, and infrastructure are utilized to cluster regions based on women’s safety. Lastly, integration with map APIs and cab service vendors addresses the travel aspect, facilitating real-time alerts for deviations into unsafe areas. Results encompass a nuanced correlation matrix classifying regions based on safety clusters, offering valuable insights for urban planning and law enforcement. Integration with cab services ensures SafeRoutes not only identifies safe paths but actively contributes to enhancing women’s safety during transit. The anticipated outcome positions SafeRoutes as a pioneering solution, contributing substantially to the discourse on urban safety and establishing a benchmark for future research.

Abstract This study attempts to accelerate the learning ability of an artificial electric field algorithm (AEFA) by attributing it with two mechanisms: elitism and opposition-based learning. Elitism advances the convergence of the AEFA towards global optima by retaining the fine-tuned solutions obtained thus far, and opposition-based learning helps enhance its exploration ability. The new version of the AEFA, called elitist opposition leaning-based AEFA (EOAEFA), retains the properties of the basic AEFA while taking advantage of both elitism and opposition-based learning. Hence, the improved version attempts to reach optimum solutions by enabling the diversification of solutions with guaranteed convergence. Higher-order neural networks (HONNs) have single-layer adjustable parameters, fast learning, a robust fault tolerance, and good approximation ability compared with multilayer neural networks. They consider a higher order of input signals, increased the dimensionality of inputs through functional expansion and could thus discriminate between them. However, determining the number of expansion units in HONNs along with their associated parameters (i.e., weight and threshold) is a bottleneck in the design of such networks. Here, we used EOAEFA to design two HONNs, namely, a pi-sigma neural network and a functional link artificial neural network, called EOAEFA-PSNN and EOAEFA-FLN, respectively, in a fully automated manner. The proposed models were evaluated on financial time-series datasets, focusing on predicting four closing prices, four exchange rates, and three energy prices. Experiments, comparative studies, and statistical tests were conducted to establish the efficacy of the proposed approach.

This study addresses the persistent challenge of in-vehicle noise, a significant factor affecting customer satisfaction and safety in the automotive industry. Despite advancements in understanding various noise sources and mitigation strategies, vehicle noise continues to contribute to driver and passenger discomfort, impacting stress levels, fatigue, and overall quality of life. Recent research has made significant strides in classifying in-vehicle noise, yet the complexity of obtaining comprehensive and diverse datasets remains a major hurdle, given the variability and transient nature of these noises. To overcome these challenges, our research introduces an innovative approach using Few-shot Learning (FSL). We propose a unique FSL model that integrates a Triplet-trained Prototypical Network for the classification of in-vehicle noises. This model is particularly adept at learning robust feature representations from limited data. The application of triplet sampling and loss significantly enhances the model’s ability to distinguish between various types of in-vehicle noises. Our methodology was rigorously tested using a specially curated dataset of in-vehicle noises, reflecting real-world diversity. The experimental results, obtained through 10-fold cross-validation, demonstrate an exceptional average accuracy of 96.81% on a 9-way 1-shot task. This level of accuracy, achieved with a limited amount of training data, not only attests to the effectiveness of our model but also marks a significant advancement in the field of acoustic classification. Our study’s findings highlight the potential of FSL in addressing complex challenges in the automotive industry, paving the way for more effective noise reduction strategies and improved vehicle design.

With the increasing reliance on technology in traffic management systems, ensuring road safety and protecting the integrity of these systems against cyber threats have become critical concerns. This research paper investigates the potential of reinforcement learning techniques in enhancing both road safety and cyber security of traffic management systems. The paper explores the theoretical foundations of reinforcement learning, discusses its applications in traffic management, and presents case studies and empirical evidence demonstrating its effectiveness in improving road safety and mitigating cyber security risks. The findings indicate that reinforcement learning can contribute to the development of intelligent and secure traffic management systems, thus minimizing accidents and protecting against cyber-attacks.

Compressive strength (CS) has been considered as the utmost critical parameter while designing concrete structures. Usually, it is determined through laboratory tests, which are expensive, time consuming, and requires consumptions of materials. Therefore, correct prediction of CS before actual placement of concrete is highly desirable. The relationship among the constituent materials that forms concrete structures is highly nonlinear and necessitates application of intelligent methods. Though a few such methods like artificial neural network (ANN) based models are available in the literature; their performance in the context is limited to certain extent and they have their own merits and demerits. Hence, to address some of the limitations of ANN (non-higher order Neural Network) based models, this contribution proposed a hybrid model in which a flat network i.e., a type of higher order neural network- functional link artificial neural network (FLANN) is used as the base structure and genetic algorithm (GA) is employed to find out the optimal FLANN parameters (i.e., GA + FLANN). The training process comprises selection of connection weights, bias, as well as optimal number of basis functions of FLANN by GA rather fixing them earlier. Thus, an optimal FLANN is crafted on fly from exploitation of training data. The proposed model is used to assess the CS of concrete cements from datasets available in the literature considering samples of curing age at 3, 7, 14, 28, 56, and 91 days. A rolling window is used for input selection and five evaluation metrics are used for performance evaluation. On an average, the GA + FLANN obtained 0.477052 MAPE (mean absolute percentage of error), 0.598067 ARV (average relative variance), 0.593862 UT (U of Theil’s statistic), 0.551452 NMSE (normalized mean squared error), and 0.206135 SD (standard deviation) values which are the lowest compared to others. The superiority of the model is established through comparative studies and statistical significance tests.

Background: Autism spectrum disorder (ASD) is a neurodevelopmental disorder exhibiting heterogeneous characteristics in patients, including variability in developmental progression and distinct neuroanatomical features influenced by sex and age. Recent advances in deep learning models based on functional connectivity (FC) graphs have produced promising results, but they have focused on generalized global activation patterns and failed to capture specialized regional characteristics and accurately assess disease indications. Methods: To overcome these limitations, we propose a novel deep learning method that models FC with multi-head attention, which enables simultaneous modeling of the intricate and variable patterns of brain connectivity associated with ASD, effectively extracting abnormal patterns of brain connectivity. The proposed method not only identifies region-specific correlations but also emphasizes connections at specific, transient time points from diverse perspectives. The extracted FC is transformed into a graph, assigning weighted labels to the edges to reflect the degree of correlation, which is then processed using a graph neural network capable of handling edge labels. Results: Experiments on the autism brain imaging data exchange (ABIDE) I and II datasets, which include a heterogeneous cohort, showed superior performance over the state-of-the-art methods, improving accuracy by up to 3.7%p. The incorporation of multi-head attention in FC analysis markedly improved the distinction between typical brains and those affected by ASD. Additionally, the ablation study validated diverse brain characteristics in ASD patients across different ages and sexes, offering insightful interpretations. Conclusion: These results emphasize the effectiveness of the method in enhancing diagnostic accuracy and its potential in advancing neurological research for ASD diagnosis.

Cryptocurrencies have carved out a significant presence in financial transactions during the past few years. Cryptocurrency market performs similarly to other financial markets with considerable nonlinearity and volatility and its prediction is a growing research area. It is challenging to capture the inherent uncertainties connected with cryptocurrency using the currently used conventional methodologies. The popularity of random vector functional link networks (RVFLN) is attributed to its simple structural layout, quick rate of learning, and enhanced generalization ability. It computes the output layer weights using non-iterative techniques like least square methods or iterative techniques like gradient methods, and assigns hidden neuron parameters at random. Random initialization of non-optimal hidden neuron settings, however, degrades the performance. Population-based metaheuristics are a superior option to random initialization for determining the ideal parameters and avoiding the problem of local optima stagnation. In the current article, an elitist artificial electric field algorithm (eAEFA) for training RVFLN is proposed. Here, eAEFA is utilized to create an ideal RVFLN by determining the weights and biases of the hidden layer connections. The elitism method is used by AEFA to maximize its strength. Here, the most suitable entities are directly inserted to create the population of the following generation. By predicting the closing values of six widely used cryptocurrencies, including Bitcoin, Litecoin, Ethereum, ZEC, XLM, and Ripple, one may determine how well the eAEFA+RVFLN model is performing. For comparison study, models including ARIMA, multi-layer perceptron (MLP), basic RVFLN, support vector regression (SVR), LSTM, GA trained RVFLN, and AEFA trained RVFLN are also constructed concurrently. In terms of performance and statistical significance testing, the suggested eAEFA+RVFLN findings outperform the comparator models. On an average, it achieves a MAPE (mean absolute percentage of error) value of 0.0573, R2 (coefficient of determination) of 0.9589, POCID (prediction of change in direction) of 0.9676, RMSE (root mean squared error) of 0.0685, MAE (mean absolute error) of 0.0727 and an average rank of 1.346; as a result, it is possible to recommend it as a useful financial forecasting tool.

This work presents a novel approach by considering teaching learning based optimization (TLBO) and radial basis function neural networks (RBFNs) for building a classifier for the databases with missing values and irrelevant features. The least square estimator and relief algorithm have been used for imputing the database and evaluating the relevance of features, respectively. The preprocessed dataset is used for developing a classifier based on TLBO trained RBFNs for generating a concise and meaningful description for each class that can be used to classify subsequent instances with no known class label. The method is evaluated extensively through a few bench-mark datasets obtained from UCI repository. The experimental results confirm that our approach can be a promising tool towards constructing a classifier from the databases with missing values and irrelevant attributes.

Chemical reaction optimization (CRO) algorithm is a robust metaheuristic by simulating the process of natural chemical reaction and has been applied for solving numerous realistic problems. However, it is suffering from low population diversity, slow convergence, and deprived from local search ability. To boost the search operation of CRO, this article proposes an improved CRO method termed as ICRO. We achieve the improvements in two phases. First, to start the search operation with a better-quality population, we propose a new method to initialize population which helps in enhancing diversification of the search space. Second, to intensify the local search ability we integrate Nelder-Mead simplex method with CRO. The ICRO is then evaluated on a few benchmark functions for optimization and found apposite. Dendritic neuron model (DNM) using additive and multiplicative-based aggregation functions has been emerging as a machine learning approach and found successful in many engineering applications. This study attempts to advance the predictive accuracy of DNM through maintaining a decent steadiness between exploitation and exploration of its search space with the proposed ICRO, termed as ICRO-DNM. The powerful global search ability of ICRO synergies with better approximation capability of DNM thus, able to overcome the limitations of conventional back propagation learning based DNM. ICRO-DNM is evaluated on closing price prediction for four stock index and net asset values of four mutual fund datasets and found efficient. The consequences of this attempt are: 1) Proposed ICRO shows robust parameter optimization ability for benchmark functions and DNM compared to basic CRO, particle swarm optimization (PSO), and genetic algorithm (GA) and 2) The learning paradigm formed due to reasonable amalgamation of ICRO and DNM is pretty able to capture the underlying uncertainties coupled with financial data, produces more precise and steady predictions and significantly different from other forecasts.

In recent industrial applications, machine learning technology is proving useful in preventing equipment failures in advance through early failure diagnosis. In particular, we show that different domains can be linked through adversarial learning with data available in different working conditions to facilitate the training of the model, as it is impractical to acquire data for all conditions in real-world applications. Nevertheless, the initial failure is a difficult problem to diagnose because it does not show a significant difference from the normal data between different conditions. Moreover, if only the domain discriminator is judged when adapting the domain, it tends to easily cause misclassification, so the reliability of the detection result needs to be improved. In this study, we propose a new learning method that improves classification performance by sharing the classification characteristics of the classifier for each task with the target domain characteristic generator. The proposed mechanism uses spatial attention to extract the focused partial information of the feature generator and discriminator, and further enhances task-specific features using the attention mechanism between the two extracted information types. Addresses the challenge of implementing both domain adaptation and classification. Extensive experimentation demonstrates efficiency and improved classification performance on benchmark and real-world application datasets. In real machine cases, the classification accuracy is improved by almost 4%. In addition, the negative impact on false alarms was lowered by increasing the classification accuracy of minimum failures. Convincingly demonstrate model effectiveness by performing an empirical analysis of the method through ablation analysis and visualization.

Prediction of currency exchange rate becomes highly desirable due to its greater role in financial and managerial decision making process. The fluctuations in exchange rate affect the economy of a country. Hence, over the years different types of neural network models along with statistical models are developed to predict the currency exchange rates of different countries with varying parameters. In this paper, we divide our effort into two parts. In first part, we have reviewed a few selected models of neural networks and statistics including fundamental and technical aspects of currency exchange rate prediction. Additionally, a thorough and careful experimental result analysis has been conducted on the models reviewed in part one. A committee machine has been proposed in part two to address the shortcomings of both neural networks and statistical models in the context of exchange rate prediction. Our study reveals that the currency exchange rates with multi-layer neural networks having Bayesian learning predictive accuracy is better than multi-layer neural networks with back-propagation learning. However, in the case of higher-order neural network multi-stage radial basis function network is predicting better than single stage radial basis function network. In the case of statistical models, it is drawn that under the umbrella of root mean square error measure, random walk is predicting better than other models of this category, whereas variance based model predicts better than rest of the models grouped under normalized mean square error measure. On the other hand, the integrated model is performing better than its counterpart like models with stand-alone mode. Moreover, our newly proposed committee machine is drawing a clear line over all the models while predicting exchange rate of GBP/USD.

In this paper, we propose a novel method of cluster analysis called unsupervised functional link artificial neural networks (UFLANNs), which inherit the best characteristics of functional link artificial neural networks and self-organizing feature maps (SOFMs). UFLANNs adopt three types of basis functions such as Chebyshev, Legendre orthogonal polynomials, and power series for mapping the input data into a new feature space with higher dimensions, where the objects are clustered based on the principle of competitive learning of SOFMs. The effectiveness of this algorithm has been tested with various artificial and real-life datasets including remote sensing images. A thorough comparison with other popular clustering algorithms shows that the proposed method is promising in revealing clusters from many complex datasets.

Urban mobility is a vital aspect of any city and often influences its physical shape as well as its level of economic and social development. A thorough analysis of mobility patterns in urban areas can provide various benefits, such as the prediction of traffic flow and public transportation usage. In particular, based on its exceptional ability to extract patterns from complex large-scale data, embedding based on deep learning is a promising method for analyzing the mobility patterns of urban residents. However, as urban mobility becomes increasingly complex, it becomes difficult to embed patterns into a single vector because of its limited capacity. In this paper, we propose a novel method for analyzing urban mobility based on deep learning. The proposed method involves clustering mobility patterns and embedding them to capture their implicit meaning. Clustering groups mobility patterns based on their spatiotemporal characteristics, and embedding provides meaningful information regarding both individual residents (i.e., personalized mobility) and all residents as a whole, enabling a more effective analysis of mobility patterns. Experiments were performed to predict the successive points of interest (POIs) based on transportation data collected from 1.5 million citizens in a large metropolitan city; the results demonstrate that the proposed method achieves top-1, 3, and 5 accuracies of 73.64%, 88.65%, and 91.54%, respectively, which are much higher than those of the conventional method (59.48%, 75.85%, and 80.1%, respectively). We also demonstrate that the proposed method facilitates the analysis of urban mobility through arithmetic operations between POI vectors.

In recent years, various deep learning models have been developed for the fault diagnosis of rotating machines. However, in practical applications related to fault diagnosis, it is difficult to immediately implement a trained model because the distribution of source data and target domain data have different distributions. Additionally, collecting failure data for various operating conditions is time consuming and expensive. In this paper, we introduce a new transformation method for the latent space between domains using the source domain and normal data of the target domain that can be easily collected. Inspired by semantic transformations in an embedded space in the field of word embedding, discrepancies between the distribution of the source and target domains are minimized by transforming the latent representation space in which fault attributes are preserved. To match the feature area and distribution, spatial attention is applied to learn the latent feature spaces, and the 1D CNN LSTM architecture is implemented to maximize the intra-class classification. The proposed model was validated for two types of rotating machines such as a dataset of rolling bearings as CWRU and a gearbox dataset of heavy machinery. Experimental results show the proposed method has higher cross-domain diagnostic accuracy than others, therefore showing reliable generalization performance in rotating machines operating under various conditions.

Predicting residential energy consumption is tantamount to forecasting a multivariate time series. A specific window for several sensor signals can induce various features extracted to forecast the energy consumption by using a prediction model. However, it is still a challenging task because of irregular patterns inside including hidden correlations between power attributes. In order to extract the complicated irregular energy patterns and selectively learn the spatiotemporal features to reduce the translational variance between energy attributes, we propose a deep learning model based on the multi-headed attention with the convolutional recurrent neural network. It exploits the attention scores calculated with softmax and dot product operation in the network to model the transient and impulsive nature of energy demand. Experiments with the dataset of University of California, Irvine (UCI) household electric power consumption consisting of a total 2,075,259 time-series show that the proposed model reduces the prediction error by 31.01% compared to the state-of-the-art deep learning model. Especially, the multi-headed attention improves the prediction performance even more by up to 27.91% than the single-attention.

Recently, the concept of Internet of Things has widely proliferated to offer advanced connectivity between devices, systems, and services that continuously obtain enormous amounts of data from sensors. Recognizing context from the sensor data plays a crucial role in adding value to the raw sensor data. In this article, we propose a context-aware system through device-oriented modeling for the Internet of Things using modular Bayesian networks based on our previous study. A Bayesian network can handle flexibly the uncertain environments of frequent changes in device configuration, and the proposed system can enable us to adjust to the changing Internet of Things environment, making it more flexible. The main contribution of the article lies in the realization of the modular context-aware system with device-oriented modeling of Bayesian networks in smart home and the verification of the usability through a subjective test with 116 people. In addition, we evaluate the performance of the proposed system and show the reduction of time complexity using the real data. Compared to other methods such as decision tree and monolithic Bayesian network, the performance improvement is statistically significant according to t-test.

Peer-to-Peer (P2P) lending transactions take place by the lenders choosing a borrower and lending money. It is important to predict whether a borrower can repay because the lenders must bear the credit risk when the borrower defaults, but it is difficult to design feature extractors with very complex information about borrowers and loan products. In this paper, we present an architecture of deep convolutional neural network (CNN) for predicting the repayment in P2P social lending to extract features automatically and improve the performance. CNN is a deep learning model for classifying complex data, which extracts discriminative features automatically by convolution operation on lending data. We classify the borrower’s loan status by capturing the robust features and learning the patterns. Experimental results with 5-fold cross-validation show that our method automatically extracts complex features and is effective in repayment prediction on Lending Club data. In comparison with other machine learning methods, the standard CNN has achieved the highest performance with 75.86%. Exploiting various CNN models such as Inception, ResNet, and Inception-ResNet results in the state-of-the-art performance of 77.78%. We also demonstrate that the features extracted by our model are better performed by projecting the samples into the feature space.

As energy demand grows globally, the energy management system (EMS) is becoming increasingly important. Energy prediction is an essential component in the first step to create a management plan in EMS. Conventional energy prediction models focus on prediction performance, but in order to build an efficient system, it is necessary to predict energy demand according to various conditions. In this paper, we propose a method to predict energy demand in various situations using a deep learning model based on an autoencoder. This model consists of a projector that defines an appropriate state for a given situation and a predictor that forecasts energy demand from the defined state. The proposed model produces consumption predictions for 15, 30, 45, and 60 min with 60-min demand to date. In the experiments with household electric power consumption data for five years, this model not only has a better performance with a mean squared error of 0.384 than the conventional models, but also improves the capacity to explain the results of prediction by visualizing the state with t-SNE algorithm. Despite unsupervised representation learning, we confirm that the proposed model defines the state well and predicts the energy demand accordingly.

Human activity recognition has been gaining more and more attention from researchers in recent years, particularly with the use of widespread and commercially available devices such as smartphones. However, most of the existing works focus on discriminative classifiers while neglecting the inherent time-series and continuous characteristics of sensor data. To address this, we propose a two-stage continuous hidden Markov model framework, which also takes advantage of the innate hierarchical structure of basic activities. This kind of system architecture not only enables the use of different feature subsets on different subclasses, which effectively reduces feature computation overhead, but also allows for varying number of states and iterations. Experiments show that the hierarchical structure dramatically increases classification performance. We analyze the behavior of the accelerometer and gyroscope signals for each activity through graphs, and with added fine tuning of states and training iterations, the proposed method is able to achieve an overall accuracy of up to 93.18%, which is the best performance among the state-of-the-art classifiers for the problem at hand.

The cyber-physical system (CPS) is a next-generation smart system that combines computing with physical space. It has been applied in various fields because the uncertainty of the physical world can be ideally controlled using cyber technology. In terms of environmental control, studies have been conducted to enhance the effectiveness of the service by inducing ideal emotions in the service space. This paper proposes a CPS control system for inducing emotion based on multiple sensors. The CPS can expand the constrained environmental sensors of the physical space variously by combining the virtual space with the physical space. The cyber space is constructed in a Unity 3D space that can be experienced through virtual reality devices. We collect the temperature, humidity, dust concentration, and current emotion in the physical space as an environmental control elements, and the control illumination, color temperature, video, sound and volume in the cyber space. The proposed system consists of an emotion prediction module using modular Bayesian networks and an optimal stimulus decision module for deriving the predicted emotion to the target emotion based on utility theory and reinforcement learning. To verify the system, the performance is evaluated using the data collected from real situations.

This work presents an accurate and smooth functional link artificial neural network (FLANN) for classification of noisy database. The accuracy and smoothness of the network is taken birth by suitably tuning the parameters of FLANN using differential evolution and filter based feature selection. We use Qclean algorithm for identification of noise, information gain theory for filtering irrelevant features, and then supplied the remaining relevant attributes to the functional expansion unit of FLANN, which in turn map lower to higher dimensional feature space for constructing a smooth and accurate classifier. In specific, the differential evolution is used to fine tune the weight vector of this network and some trigonometric functions are used in functional expansion unit. The proposed approach is validated with a few benchmarking highly skewed and balanced dataset retrieved from University of California, Irvine (UCI) repository with a range of 5-20% noise. The insightful experimental study signifies the propensity of noise in the classification accuracy of a database with a range of noise from 5-20%. Moreover, our method suggests that noisy samples along with irrelevant set of attributes are deceptive and weakening the reliability of the classifier, therefore, it is required to reduce its effect before or during the process of classification.

Recently, recognizing a user’s daily activity using a smartphone and wearable sensors has become a popular issue. However, in contrast with the ideal definition of an experiment, there could be numerous complex activities in real life with respect to its various background and contexts: time, space, age, culture, and so on. Recognizing these complex activities with limited low-power sensors, considering the power and memory constraints of the wearable environment and the user’s obtrusiveness at once is not an easy problem, although it is very crucial for the activity recognizer to be practically useful. In this paper, we recognize activity of eating, which is one of the most typical examples of a complex activity, using only daily low-power mobile and wearable sensors. To organize the related contexts systemically, we have constructed the context model based on activity theory and the “Five W’s”, and propose a Bayesian network with 88 nodes to predict uncertain contexts probabilistically. The structure of the proposed Bayesian network is designed by a modular and tree-structured approach to reduce the time complexity and increase the scalability. To evaluate the proposed method, we collected the data with 10 different activities from 25 volunteers of various ages, occupations, and jobs, and have obtained 79.71% accuracy, which outperforms other conventional classifiers by 7.54–14.4%. Analyses of the results showed that our probabilistic approach could also give approximate results even when one of contexts or sensor values has a very heterogeneous pattern or is missing.

An office delivery robot receives a large amount of sensory data and there is uncertainty in its action outcomes. The robot should not only accomplish its goals using environmental information, but also consider various exceptions simultaneously. In this paper, we propose a hybrid system using hierarchical planning of modular behaviour selection networks to generate autonomous behaviour in the office delivery robot. Behaviour selection networks, one of the well-known behaviour-based methods suitable for goal-oriented tasks, are made up of several smaller behaviour modules. Planning is attached to the construct and adjust sequences of the modules by considering the sub-goals, the priority in each task and the user feedback. This helps the robot to quickly react in dynamic situations as well as achieve global goals efficiently. The proposed system is verified with both the Webot simulator and a Khepera II robot that runs in a real office environment carrying out delivery tasks. Experimental results have shown that a robot can achieve goals and generate module sequences successfully even in unpredictable situations. Additionally, the proposed planning method reduced the elapsed time during tasks by 17.5% since it adjusts the behaviour module sequences more effectively.

With the recent proliferation of smart phones, they become useful tools to implement high-confidence cyber-physical systems. Among many applications, context sharing systems in mobile environment attract attention with the popularization of social media. Mobile context sharing systems can share more information than web-based social network services because they can use a variety of information from mobile sensors. To share high-level contexts such as activity, emotion, and user relationship, a user had to annotate them manually in previous works. This paper proposes a mobile context sharing system that can recognize high-level contexts automatically by using Bayesian networks based on mobile logs. We have developed a ContextViewer application which consists of a phonebook and a map browser to show the feasibility of the system. Experiments of evaluating Bayesian networks and performing the SUS test confirm that the proposed system is useful.