Search

The increasing popularity of ambient assisted living solutions is claiming adaptive and scalable tools to monitor activities of daily living. Currently, most sensor-based activity recognition techniques rely on supervised learning algorithms. However, the acquisition of comprehensive training sets of activities in smart homes is expensive and violates the individual’s privacy. In this work, we address this problem by proposing a novel hybrid approach that couples collaborative active learning with probabilistic and knowledge-based reasoning. The rationale of our approach is that a generic, and possibly incomplete, knowledge-based model of activities can be refined to target specific individuals and environments by collaboratively acquiring feedback from inhabitants. Specifically, we propose a collaborative active learning method exploiting users’ feedback to (i) refine correlations among sensor events and activity types that are initially extracted from a high-level ontology, and (ii) mine temporal patterns of sensor events that are frequently generated by the execution of specific activities. A Markov Logic Network is used to recognize activities with probabilistic rules that capture both the ontological knowledge and the information obtained by active learning. We experimented our solution with a real-world dataset of activities carried out by several individuals in an interleaved fashion. Experimental results show that our collaborative and personalized active learning solution significantly improves recognition rates, while triggering a small number of feedback requests. Moreover, the overall recognition rates compare favorably with existing supervised and unsupervised activity recognition methods.

Recognizing the activities of daily living (ADLs) in multi-inhabitant settings is a challenging task. One of the major challenges is the so-called data association problem: how to assign to each user the environmental sensor events that he/she actually triggered? In this paper, we tackle this problem with a contextaware approach. Each user in the home wears a smartwatch, which is used to gather several high-level context information, like the location in the home (thanks to a micro-localization infrastructure) and the posture (e.g., sitting or standing). Context data is used to associate sensor events to the users which more likely triggered them. We show the impact of context reasoning in our framework on a dataset where up to 4 subjects perform ADLs at the same time (collaboratively or individually). We also report our experience and the lessons learned in deploying a running prototype of our method.

Semi-supervised approaches for activity recognition are a promising way to address the labeled data scarcity problem. Those methods only require a small training set in order to be initialized, and the model is continuously updated and improved over time. Among the several solutions existing in the literature, active learning is emerging as an effective technique to significantly boost the recognition rate: when the model is uncertain about the current activity performed by the user, the system asks her to provide the ground truth. This feedback is then used to update the recognition model. While active learning has been mostly proposed in single-inhabitant settings, several questions arise when such a system has to be implemented in a realistic environment with multiple users. Who to ask a feedback when the system is uncertain about a collaborative activity? In this paper, we investigate this and more questions on this topic, proposing a preliminary study of the requirements of an active learning interface for multi-inhabitant settings. In particular, we formalize the problem and we describe the solutions adopted in our system prototype.

Providing city navigation instructions to people with motion disabilities requires the knowledge of urban features like curb ramps, steps or other obstacles along the way. Since these urban features are not available from maps and change in time, crowdsourcing this information from end-users is a scalable and promising solution. Our preliminary study on wheelchair users shows that an automatic crowdsourcing mechanism is needed, avoiding users’ involvement.In this contribution we present a solution to crowdsource urban features from inertial sensors installed on a wheelchair. Activity recognition techniques based on decision trees are used to process the sensors data stream. Experimental results, conducted with data acquired from 10 real wheelchair users navigating in an outdoor environment show that our solution is effective in detecting urban features with precision around 0.9, while it is less reliable when classifying some fine-grained urban feature characteristics, like a step height. The experimental results also present our investigation aimed at identifying the best parameters for the given problem, which include number, position and type of inertial sensors, classifier type, segmentation parameters, etc.

Activity recognition on mobile device sensor data has been an active research area in mobile and pervasive computing for several years. While the majority of the proposed techniques are based on supervised learning, semi-supervised approaches are being considered to reduce the size of the training set required to initialize the model. These approaches usually apply self-training or active learning to incrementally refine the model, but their effectiveness seems to be limited to a restricted set of physical activities. We claim that the context which surrounds the user (e.g., time, location, proximity to transportation routes) combined with common knowledge about the relationship between context and human activities could be effective in significantly increasing the set of recognized activities including those that are difficult to discriminate only considering inertial sensors, and the highly context-dependent ones. In this paper, we propose CAVIAR, a novel hybrid semi-supervised and knowledge-based system for real-time activity recognition. Our method applies semantic reasoning on context-data to refine the predictions of an incremental classifier. The recognition model is continuously updated using active learning. Results on a real dataset obtained from 26 subjects show the effectiveness of our approach in increasing the recognition rate, extending the number of recognizable activities and, most importantly, reducing the number of queries triggered by active learning. In order to evaluate the impact of context reasoning, we also compare CAVIAR with a purely statistical version, considering features computed on context-data as part of the machine learning process.

People with mobility impairments have heterogeneous needs and abilities while moving in an urban environment and hence they require personalized navigation instructions. Providing these instructions requires the knowledge of urban features like curb ramps, steps or other obstacles along the way. Since these urban features are not available from maps and change in time, crowdsourcing this information from end-users is a scalable and promising solution. However, it is inconvenient for wheelchair users to input data while on the move. Hence, an automatic crowdsourcing mechanism is needed. In this contribution we present SmartWheels, a solution to detect urban features by analyzing inertial sensors data produced by wheelchair movements. Activity recognition techniques are used to process the sensors data stream. SmartWheels is evaluated on data collected from 17 real wheelchair users navigating in a controlled environment (10 users) and in-the-wild (7 users). Experimental results show that SmartWheels is a viable solution to detect urban features, in particular by applying specific strategies based on the confidence assigned to predictions by the classifier.

Smart-homectivity recognition is an enabling tool for a wide range of ambient assisted living applications. The recognition of ADLs usually relies on supervised learning or knowledge-based reasoning techniques. In order to overcome the well-known limitations of those two approaches and, at the same time, to combine their strengths to improve the recognition rate, many researchers investigated Markov Logic Networks (MLNs). However, MLNs require a non-trivial effort by experts to properly model probabilities in terms of weights. In this paper, we propose a novel method based on ProbLog. ProbLog is a probabilistic extension of Prolog, which allows to explicitly define probabilistic facts and rules. With respect to MLN, the inference mode of ProbLog is based on the closed-world assumption and it has faster response times. We propose a simple and flexible ProbLog model, which we exploit to recognize complex ADLs in an online fashion. Considering a dataset with 21 subjects, our results show that our method reaches high F-measure (83%). Moreover, we also show that the response time of ProbLog is satisfying for real-time applications.

Due to the emerging popularity of pervasive healthcare applications, tools for monitoring activities in smart homes are gaining momentum. Existing methods mainly rely on supervised learning algorithms for recognizing activities based on sensor data. A key issue with those approaches is the acquisition of comprehensive training sets of activities. Indeed, that task incurs significant costs in terms of manual labeling effort; moreover, labeling by external observers violates the individual's privacy. For these reasons, there is an increasing interest in unsupervised activity recognition methods. A popular approach relies on knowledge-based models expressed by ontologies of activities, environment and sensors. Unfortunately, those models require significant knowledge engineering efforts, and are often limited to a specific application. In this paper, we address the issues of existing methods by proposing a novel hybrid approach. Our intuition is that a generic knowledge-based model of activities can be refined to target specific individuals and environments by collaboratively acquiring feedback from inhabitants. Specifically, we propose a collaborative active learning method to refine correlations among sensor events and activity types that are initially extracted from a high-level ontology. Generic correlations are personalized to each target smart-home considering the similarity between the feedback target and the feedback provider in terms of environment and inhabitant's profiles. Moreover, thanks to this method, new sensors installed in the home are seamlessly integrated in the recognition framework. In order to reduce the burden of providing feedback, we also propose a technique to carefully select the conditions that trigger a feedback request. We conducted experiments with a real-world dataset and a generic ontology of activities. Results show that our hybrid method outperforms state-of-the-art supervised and unsupervised activity recognition techniques while triggering an acceptable number of feedback queries.

Although the method discussed in this work already obtained promising results, we plan to extend it in several directions. First of all, anomaly recognition relies on a too rigid formalisms which is not able to capture the intrinsic noise of sensor measurements. Moreover, semi-supervised ADLs recognition techniques (e.g., active learning) will be evaluated in order to better adapt the model to different users and environments. Last but not least, we aim to improve our algorithms in order to support parallel activities and multi-inhabitants scenarios.

Recognition of activities of daily living (ADLs) performed in smart homes proved to be very effective when the interaction of the inhabitant with household items is considered. Analyzing how objects are manipulated can be particularly useful, in combination with other sensor data, to detect anomalies in performing ADLs, and hence to support early diagnosis of cognitive impairments for elderly people. Recent improvements in sensing technologies can overcome several limitations of the existing techniques to detect object manipulations, often based on RFID, wearable sensors and/or computer vision methods. In this work we propose an unobtrusive solution which shifts all the monitoring burden at the objects side. In particular, we investigate the effectiveness of using tiny BLE beacons equipped with accelerometer and temperature sensors attached to everyday objects. We adopt statistical methods to analyze in realtime the accelerometer data coming from the objects, with the purpose of detecting specific manipulations performed by seniors in their homes. We describe our technique and we present the preliminary results obtained by evaluating the method on a real dataset. The results indicate the potential utility of the method in enriching ADLs and abnormal behaviors recognition systems, by providing detailed information about object manipulations.

Recognition of activities of daily living (ADLs) is an enabling technology for several ubiquitous computing applications. In this field, most activity recognition systems rely on supervised learning methods to extract activity models from labeled datasets. An inherent problem of that approach consists in the acquisition of comprehensive activity datasets, which is expensive and may violate individuals' privacy. The problem is particularly challenging when focusing on complex ADLs, which are characterized by large intra- and inter-personal variability of execution. In this paper, we propose an unsupervised method to recognize complex ADLs exploiting the semantics of activities, context data, and sensing devices. Through ontological reasoning, we derive semantic correlations among activities and sensor events. By matching observed sensor events with semantic correlations, a statistical reasoner formulates initial hypotheses about the occurred activities. Those hypotheses are refined through probabilistic reasoning, exploiting semantic constraints derived from the ontology. Extensive experiments with real-world datasets show that the accuracy of our unsupervised method is comparable to the one of state of the art supervised approaches.

Several researchers have proposed methods and designed systems for the automatic recognition of activities and abnormal behaviors with the goal of early detecting cognitive impairment. In this paper, we propose LOTAR, a hybrid behavioral analysis system coupling state of the art machine learning techniques with knowledge-based and data mining methods. Medical models designed in collaboration with cognitive neuroscience researchers guide the recognition of short- and long-term abnormal behaviors. In particular, we focus on historical behavior analysis for long-term anomaly detection, which is the principal novelty with respect to our previous works. We present preliminary results obtained by evaluating the method on a dataset acquired during three months of experimentation in a real patient's home. Results indicate the potential utility of the system for long-term monitoring of cognitive health.

Objective In an ageing world population more citizens are at risk of cognitive impairment, with negative consequences on their ability of independent living, quality of life and sustainability of healthcare systems. Cognitive neuroscience researchers have identified behavioral anomalies that are significant indicators of cognitive decline. A general goal is the design of innovative methods and tools for continuously monitoring the functional abilities of the seniors at risk and reporting the behavioral anomalies to the clinicians. SmartFABER is a pervasive system targeting this objective. Methods A non-intrusive sensor network continuously acquires data about the interaction of the senior with the home environment during daily activities. A novel hybrid statistical and knowledge-based technique is used to analyses this data and detect the behavioral anomalies, whose history is presented through a dashboard to the clinicians. Differently from related works, SmartFABER can detect abnormal behaviors at a fine-grained level. Results We have fully implemented the system and evaluated it using real datasets, partly generated by performing activities in a smart home laboratory, and partly acquired during several months of monitoring of the instrumented home of a senior diagnosed with MCI. Experimental results, including comparisons with other activity recognition techniques, show the effectiveness of SmartFABER in terms of recognition rates.