A Curiosity Estimation in Storytelling with Picture Books for Children Using Wearable Sensors.

Storytelling is one of the most important learning activities for children since reading aloud from a picture book stimulates children's curiosity, emotional development, and imagination. For effective education, the procedures for storytelling activities need to be improved according to the children's level of curiosity. However, young children are not able to complete questionnaires, making it difficult to analyze their level of interest. This paper proposes a method to estimate children's curiosity in picture book reading activities at five levels by recognizing children's behavior using acceleration and angular velocity sensors placed on their heads. We investigated the relationship between children's behaviors and their levels of curiosity, listed all observed behaviors, and clarified the behavior for estimating curiosity. Furthermore, we conducted experiments using motion sensors to estimate these behaviors and confirmed that the accuracy of estimating curiosity from sensor data is approximately 72%.

A Study on the Influence of Sensors in Frequency and Time Domains on Context Recognition.

Adaptive AI for context and activity recognition remains a relatively unexplored field due to difficulty in collecting sufficient information to develop supervised models. Additionally, building a dataset for human context activities "in the wild" demands time and human resources, which explains the lack of public datasets available. Some of the available datasets for activity recognition were collected using wearable sensors, since they are less invasive than images and precisely capture a user's movements in time series. However, frequency series contain more information about sensors' signals. In this paper, we investigate the use of feature engineering to improve the performance of a Deep Learning model. Thus, we propose using Fast Fourier Transform algorithms to extract features from frequency series instead of time series. We evaluated our approach on the ExtraSensory and WISDM datasets. The results show that using Fast Fourier Transform algorithms to extract features performed better than using statistics measures to extract features from temporal series. Additionally, we examined the impact of individual sensors on identifying specific labels and proved that incorporating more sensors enhances the model's effectiveness. On the ExtraSensory dataset, the use of frequency features outperformed that of time-domain features by 8.9 p.p., 0.2 p.p., 39.5 p.p., and 0.4 p.p. in Standing, Sitting, Lying Down, and Walking activities, respectively, and on the WISDM dataset, the model performance improved by 1.7 p.p., just by using feature engineering.

A General Framework for Making Context-Recognition Systems More Energy Efficient.

Context recognition using wearable devices is a mature research area, but one of the biggest issues it faces is the high energy consumption of the device that is sensing and processing the data. In this work we propose three different methods for optimizing its energy use. We also show how to combine all three methods to further increase the energy savings. The methods work by adapting system settings (sensors used, sampling frequency, duty cycling, etc.) to both the detected context and directly to the sensor data. This is done by mathematically modeling the influence of different system settings and using multiobjective optimization to find the best ones. The proposed methodology is tested on four different context-recognition tasks where we show that it can generate accurate energy-efficient solutions-in one case reducing energy consumption by 95% in exchange for only four percentage points of accuracy. We also show that the method is general, requires next to no expert knowledge about the domain being optimized, and that it outperforms two approaches from the related work.

New Position Candidate Identification via Clustering toward an Extensible On-Body Smartphone Localization System.

On-body device position awareness plays an important role in providing smartphone-based services with high levels of usability and quality. Traditionally, the problem assumed that the positions that were supported by the system were fixed at the time of design. Thus, if a user stores his/her terminal into an unsupported position, the system forcibly classifies it into one of the supported positions. In contrast, we propose a framework to discover new positions that are not initially supported by the system, which adds them as recognition targets via labeling by a user and re-training on-the-fly. In this article, we focus on a component of identifying a set of samples that are derived from a single storing position, which we call new position candidate identification. Clustering is applied as a key component to prepare a reliable dataset for re-training and to reduce the user's burden of labeling. Specifically, density-based spatial clustering of applications with noise (DBSCAN) is employed because it does not require the number of clusters in advance. We propose a method of finding an optimal value of a main parameter, Eps-neighborhood (eps), which affects the accuracy of the resultant clusters. Simulation-based experiments show that the proposed method performs as if the number of new positions were known in advance. Furthermore, we clarify the timing of performing the new position candidate identification process, in which we propose criteria for qualifying a cluster as the one comprising a new position.

Integrating Multiple Models Using Image-as-Documents Approach for Recognizing Fine-Grained Home Contexts.

To implement fine-grained context recognition that is accurate and affordable for general households, we present a novel technique that integrates multiple image-based cognitive APIs and light-weight machine learning. Our key idea is to regard every image as a document by exploiting "tags" derived by multiple APIs. The aim of this paper is to compare API-based models' performance and improve the recognition accuracy by preserving the affordability for general households. We present a novel method for further improving the recognition accuracy based on multiple cognitive APIs and four modules, fork integration, majority voting, score voting, and range voting.

Evaluation on Context Recognition Using Temperature Sensors in the Nostrils.

We can benefit from various services with context recognition using wearable sensors. In this study, we focus on the contexts acquired from sensor data in the nostrils. Nostrils can provide various contexts on breathing, nasal congestion, and higher level contexts including psychological and health states. In this paper, we propose a context recognition method using the information in the nostril. We develop a system to acquire the temperature in the nostrils using small temperature sensors connected to glasses. As a result of the evaluations, the proposed system can detect breathing correctly, workload at an accuracy of 96.4%, six behaviors at an accuracy of 54%, and eight behaviors in daily life at an accuracy of 86%. Moreover, the proposed system can detect nasal congestion, therefore, it can log nasal cycles that are considered to have a relationship with the autonomic nerves and/or biological states.

Context Mining of Sedentary Behaviour for Promoting Self-Awareness Using a Smartphone.

Sedentary behaviour is increasing due to societal changes and is related to prolonged periods of sitting. There is sufficient evidence proving that sedentary behaviour has a negative impact on people's health and wellness. This paper presents our research findings on how to mine the temporal contexts of sedentary behaviour by utilizing the on-board sensors of a smartphone. We use the accelerometer sensor of the smartphone to recognize user situations (i.e., still or active). If our model confirms that the user context is still, then there is a high probability of being sedentary. Then, we process the environmental sound to recognize the micro-context, such as working on a computer or watching television during leisure time. Our goal is to reduce sedentary behaviour by suggesting preventive interventions to take short breaks during prolonged sitting to be more active. We achieve this goal by providing the visualization to the user, who wants to monitor his/her sedentary behaviour to reduce unhealthy routines for self-management purposes. The main contribution of this paper is two-fold: (i) an initial implementation of the proposed framework supporting real-time context identification; (ii) testing and evaluation of the framework, which suggest that our application is capable of substantially reducing sedentary behaviour and assisting users to be active.

Using Markov Chains and Multi-Objective Optimization for Energy-Efficient Context Recognition.

The recognition of the user's context with wearable sensing systems is a common problem in ubiquitous computing. However, the typically small battery of such systems often makes continuous recognition impractical. The strain on the battery can be reduced if the sensor setting is adapted to each context. We propose a method that efficiently finds near-optimal sensor settings for each context. It uses Markov chains to simulate the behavior of the system in different configurations and the multi-objective genetic algorithm to find a set of good non-dominated configurations. The method was evaluated on three real-life datasets and found good trade-offs between the system's energy expenditure and the system's accuracy. One of the solutions, for example, consumed five-times less energy than the default one, while sacrificing only two percentage points of accuracy.

A Context-Recognition-Aided PDR Localization Method Based on the Hidden Markov Model.

Indoor positioning has recently become an important field of interest because global navigation satellite systems (GNSS) are usually unavailable in indoor environments. Pedestrian dead reckoning (PDR) is a promising localization technique for indoor environments since it can be implemented on widely used smartphones equipped with low cost inertial sensors. However, the PDR localization severely suffers from the accumulation of positioning errors, and other external calibration sources should be used. In this paper, a context-recognition-aided PDR localization model is proposed to calibrate PDR. The context is detected by employing particular human actions or characteristic objects and it is matched to the context pre-stored offline in the database to get the pedestrian's location. The Hidden Markov Model (HMM) and Recursive Viterbi Algorithm are used to do the matching, which reduces the time complexity and saves the storage. In addition, the authors design the turn detection algorithm and take the context of corner as an example to illustrate and verify the proposed model. The experimental results show that the proposed localization method can fix the pedestrian's starting point quickly and improves the positioning accuracy of PDR by 40.56% at most with perfect stability and robustness at the same time.

Ontology-Based High-Level Context Inference for Human Behavior Identification.

Recent years have witnessed a huge progress in the automatic identification of individual primitives of human behavior, such as activities or locations. However, the complex nature of human behavior demands more abstract contextual information for its analysis. This work presents an ontology-based method that combines low-level primitives of behavior, namely activity, locations and emotions, unprecedented to date, to intelligently derive more meaningful high-level context information. The paper contributes with a new open ontology describing both low-level and high-level context information, as well as their relationships. Furthermore, a framework building on the developed ontology and reasoning models is presented and evaluated. The proposed method proves to be robust while identifying high-level contexts even in the event of erroneously-detected low-level contexts. Despite reasonable inference times being obtained for a relevant set of users and instances, additional work is required to scale to long-term scenarios with a large number of users.

Service Robots for Context Recognition in Smart Environments.

BACKGROUND: A new concept for context recognition involves a smart home system, wearables, and service robots. OBJECTIVES: To find the best suited robot platform for the given use case. METHODS: An assessment of available service robots based on various factors is carried out. RESULTS: The OpenBot is best suited for the further development of this concept. CONCLUSION: The new concept will increase the precision and reliability in context recognition.

A survey of context recognition in surgery.

With the introduction of operating rooms of the future context awareness has gained importance in the surgical environment. This paper organizes and reviews different approaches for recognition of context in surgery. Major electronic research databases were queried to obtain relevant publications submitted between the years 2010 and 2015. Three different types of context were identified: (i) the surgical workflow context, (ii) surgeon's cognitive and (iii) technical state context. A total of 52 relevant studies were identified and grouped based on the type of context detected and sensors used. Different approaches were summarized to provide recommendations for future research. There is still room for improvement in terms of methods used and evaluations performed. Machine learning should be used more extensively to uncover hidden relationships between different properties of the surgeon's state, particularly when performing cognitive context recognition. Furthermore, validation protocols should be improved by performing more evaluations in situ and with a higher number of unique participants. The paper also provides a structured outline of recent context recognition methods to facilitate development of new generation context-aware surgical support systems.

Cardiorespiratory fitness estimation in free-living using wearable sensors.

OBJECTIVE: In this paper we propose artificial intelligence methods to estimate cardiorespiratory fitness (CRF) in free-living using wearable sensor data. METHODS: Our methods rely on a computational framework able to contextualize heart rate (HR) in free-living, and use context-specific HR as predictor of CRF without need for laboratory tests. In particular, we propose three estimation steps. Initially, we recognize activity primitives using accelerometer and location data. Using topic models, we group activity primitives and derive activities composites. We subsequently rank activity composites, and analyze the relation between ranked activity composites and CRF across individuals. Finally, HR data in specific activity primitives and composites is used as predictor in a hierarchical Bayesian regression model to estimate CRF level from the participant's habitual behavior in free-living. RESULTS: We show that by combining activity primitives and activity composites the proposed framework can adapt to the user and context, and outperforms other CRF estimation models, reducing estimation error between 10.3% and 22.6% on a study population of 46 participants. CONCLUSIONS: Our investigation showed that HR can be contextualized in free-living using activity primitives and activity composites and robust CRF estimation in free-living is feasible.

Cardiorespiratory fitness estimation using wearable sensors: Laboratory and free-living analysis of context-specific submaximal heart rates.

In this work, we propose to use pattern recognition methods to determine submaximal heart rate (HR) during specific contexts, such as walking at a certain speed, using wearable sensors in free living, and using context-specific HR to estimate cardiorespiratory fitness (CRF). CRF of 51 participants was assessed by a maximal exertion test (V̇o2 max). Participants wore a combined accelerometer and HR monitor during a laboratory-based simulation of activities of daily living and for 2 wk in free living. Anthropometrics, HR while lying down, and walking at predefined speeds in laboratory settings were used to estimate CRF. Explained variance (R(2)) was 0.64 for anthropometrics, and increased up to 0.74 for context-specific HR (0.73-0.78 when including fat-free mass). Next, we developed activity recognition and walking speed estimation algorithms to determine the same contexts (i.e., lying down and walking) in free living. Context-specific HR in free living was highly correlated with laboratory measurements (Pearson's r = 0.71-0.75). R(2) for CRF estimation was 0.65 when anthropometrics were used as predictors, and increased up to 0.77 when including free-living context-specific HR (i.e., HR while walking at 5.5 km/h). R(2) varied between 0.73 and 0.80 when including fat-free mass among the predictors. Root mean-square error was reduced from 354.7 to 281.0 ml/min by the inclusion of context-specific HR parameters (21% error reduction). We conclude that pattern recognition techniques can be used to contextualize HR in free living and estimated CRF with accuracy comparable to what can be obtained with laboratory measurements of HR response to walking.