Recommender systems have become critical tools for e-commerce businesses in recent years and online travel platforms such as Expedia Group have made heavy use of them in production. Contemporary travel platforms benefit greatly from the use of recommender systems as very often the space of products (trips) is quite large and shopping cycles often stretch into the weeks. Expedia Group has successfully trained and deployed multiple types of recommender systems in order to help our travelers find the perfect destination and property for them. In recommender systems literature, much attention is paid to the mathematical aspects of the field but here we focus on best practices in applying recommender systems in large-scale e-commerce for the improvement of browsing and shopping experiences. In this paper, we describe how we personalize the user experience on a number of our core pages by exploiting existing internal recommender systems and relevant recommender system literature. Additionally we note several critical lessons learned about the importance of a robust machine learning platform, the need to apply engineering best practices and how best to integrate and test recommender systems in production.

Nighttime agitation, sleep disturbances, and urinary incontinence (UI) occur frequently in individuals with dementia and can add additional burden to family caregivers, although the co-occurrence of these symptoms is not well understood. The purpose of the current study was to determine the feasibility and acceptability of using passive body sensors in community-dwelling individuals with Alzheimer's disease (AD) by family caregivers and the correlates among these distressing symptoms. A single-group, descriptive design with convenience sampling of participants with AD and their family caregivers was undertaken to address the study aims. Results showed that using body sensors was feasible and acceptable and that patterns of nocturnal agitation, sleep, and UI could be determined and were correlated in study participants. Using data from body sensors may be useful to develop and implement targeted, individualized interventions to lessen these distressing symptoms and decrease caregiver burden. Further study in this field is warranted. [ Journal of Gerontological Nursing, 44 (8), 19–26.]

1 I o h m f a k s h L p th c I p r c in 2 F a c f s F Dem Everyd Chungkuk Y KAIST 1. INTROD Imagine you c objects using hardware. To t mobile platform fingertip locali as TV screen, kitchen counte surficial intera hardware not c LED, pair of position on a su he 3-D localiz camera of a com In this demo, principle of ref reflection caste challenges. Al nteractive appl

Driven to create intuitive computing interfaces throughout our everyday space, various state-of-the-art technologies have been proposed for near-surface localization of a user's finger input such as hover or touch. However, these works require specialized hardware not commonly available, limiting the adoption of such technologies. We present SymmetriSense, a technology enabling near-surface 3-dimensional fingertip localization above arbitrary glossy surfaces using a single commodity camera device such as a smartphone. SymmetriSense addresses the localization challenges in using a single regular camera by a novel technique utilizing the principle of reflection symmetry and the fingertip's natural reflection casted upon surfaces like mirrors, granite countertops, or televisions. SymmetriSense achieves typical accuracies at sub-centimeter levels in our localization tests with dozens of volunteers and remains accurate under various environmental conditions. We hope SymmetriSense provides a technical foundation on which various everyday near-surface interactivity can be designed.

Patients with Alzheimer's Disease (AD) often experience urinary incontinence and agitation during sleep. There is some evidence that these phenomena are related, but the relationships (and the subsequent opportunity for caregiver intervention) has never been formally studied. In this work, the relationships among the times of occurrence of nighttime agitation, sleep continuity and duration, and urinary incontinence are identified for persons with AD by using innovative, non-invasive technology. Deployments in 12 homes demonstrate both the utility of the technical monitoring system and the discovered correlations between agitation and incontinence for these 12 AD patients. Implications of possible interventions are discussed. Lessons learned for technical, non-technical and health care implications are presented.

Advances in wireless sensor networks have enabled the monitoring of daily activities of elderly people. The goal of these monitoring applications is to learn normal behavior in terms of daily activities and look for any deviation, i.e., Anomalies, so that alerts can be sent to relatives or caregivers. However, human behavior is very complex, and many existing anomaly detection systems are too simplistic which cause many false alarms, resulting in unreliable systems. We present Holmes, a comprehensive anomaly detection system for daily in-home activities. Holmes accurately learns a resident's normal behavior by considering variability in daily activities based not only on a per day basis, but also considering specific days of the week, different time periods such as per week and per month, and collective, temporal, and correlation based features. This approach of learning complicated normal behaviors reduces false alarms. Also, based on resident and expert feedback, Holmes learns semantic rules that explain specific variations of activities in specific scenarios to further reduce false alarms. We evaluate Holmes using data collected from our own deployed system, public data sets, and data collected by a senior safety system provider company from an elderly resident's home. Our evaluation shows that compared to state of the art systems, Holmes reduces false positives and false negatives by at least 46% and 27%, respectively.

Home health care sensing systems are projected to streamline the efficiency of the practice of medicine by decreasing the costs of senior care and by providing preventative care to keep people out of hospitals and nursing homes. Many current sensing systems are not yet flexible enough to easily handle widely different medical applications. Empath2 provides a flexible three layer architecture that uses the Cloud and can easily be instantiated for different home health care applications. To demonstrate the flexibility of the architecture, Empath2 was instantiated for three widely different purposes. We present the design of Empath2 stressing properties that support flexibility and discuss its differences from other flexible home monitoring architectures. Evaluations for three sets of real home deployments (two of which with actual patients, and one with healthy people) are presented showing the short deployment times, short software development times, and its effectiveness for the applications at hand. Lessons learned are also presented.

We present Vocal-Diary, a voice command based ground truth collection system that uses grammar based commands from residents to log start and end of activities. Vocal-Diary ensures robustness in the presence of sounds from different environmental noise and day-to-day conversation by using two-way acknowledgement and integrating speaker recognition in the pipeline. Vocal-Diary also utilizes the sensor data produced by the underlying activity recognition system to query residents periodically to check if they forgot to log any activity. Evaluation shows that Vocal-Diary increases precision by at least 40% and recall by at least 10% compared to a system that uses voice command recognition without any acknowledgement and speaker recognition.

Home monitoring systems currently gather information about peoples activities of daily living and information regarding emergencies, however they currently lack the ability to track speech. Practical speech analysis solutions are needed to help monitor ongoing conditions such as depression, as the amount of social interaction and vocal affect is important for assessing mood and well-being. Although there are existing solutions that classify the identity and the mood of a speaker, when the acoustic signals are captured in reverberant environments they perform poorly. In this paper, we present a practical reverberation compensation method called RESONATE, which uses simulated room impulse responses to adapt a training corpus for use in multiple real reverberant rooms. We demonstrate that the system creates robust classifiers that perform within 5 -- 10% of baseline accuracy of non-reverberant environments. We demonstrate and evaluate the performance of this matched condition strategy using a public dataset, and also in controlled experiments with six rooms, and two long-term and uncontrolled real deployments. We offer a practical implementation that performs collection, feature extraction, and classification on-node, and training and simulation of training sets on a base station or cloud service.

Modeling and simulation play important roles in engineering research and design. These techniques are especially helpful in the early phases where limited detail is available about the design and where design changes are less costly. In addition, high-fidelity models can be employed at the later stages to complement testing. Models are also important research tools for understanding complex phenomena.

Auditeur is a general-purpose, energy-efficient, and context-aware acoustic event detection platform for smartphones. It enables app developers to have their app register for and get notified on a wide variety of acoustic events. Auditeur is backed by a cloud service to store user contributed sound clips and to generate an energy-efficient and context-aware classification plan for the phone. When an acoustic event type has been registered, the smartphone instantiates the necessary acoustic processing modules and wires them together to execute the plan. The phone then captures, processes, and classifies acoustic events locally and efficiently. Our analysis on user-contributed empirical data shows that Auditeur's energy-aware acoustic feature selection algorithm is capable of increasing the device lifetime by 33.4%, sacrificing less than 2% of the maximum achievable accuracy. We implement seven apps with Auditeur, and deploy them in real-world scenarios to demonstrate that Auditeur is versatile, 11.04% - 441.42% less power hungry, and 10.71% - 13.86% more accurate in detecting acoustic events, compared to state-of-the-art techniques. We present a user study to demonstrate that novice programmers can implement the core logic of interesting apps with Auditeur in less than 30 minutes, using only 15 - 20 lines of Java code.

Due to limited processing capability, contemporary smartphones cannot extract frequency domain acoustic features in real-time on the device when the sampling rate is high. We propose a solution to this problem which exploits the sparseness in speech to extract frequency domain acoustic features inside a smartphone in real-time, without requiring any support from a remote server even when the sampling rate is as high as 44.1 KHz. We perform an empirical study to quantify the sparseness in speech recorded on a smartphone and use it to obtain a highly accurate and sparse approximation of a widely used feature of speech called the Mel-Frequency Cepstral Coefficients (MFCC) efficiently. We name the new feature the sparse MFCC or sMFCC, in short. We experimentally determine the trade-offs between the approximation error and the expected speedup of sMFCC. We implement a simple spoken word recognition application using both MFCC and sMFCC features, show that sMFCC is expected to be up to 5.84 times faster and its accuracy is within 1.1% -- 3.9% of that of MFCC, and determine the conditions under which sMFCC runs in real-time.

This chapter presents a sleep monitoring system based on WISP tags. The authors show that their system accurately infers fine-grained body positions from accelerometer data collected from the WISP tags attached to the sides of a bed. Movements, duration, and bed entrances and exits are also detected by the system. The chapter presents the results of an empirical study from 10 subjects on three different mattresses in controlled experiments to show the accuracy of the inference algorithms. The authors also evaluate the accuracy of the movement detection and body position inference for six nights on one subject, and compare these results with two baseline systems. Preliminary data investigating the correlation between sleep stages from the Zeo and movement is also presented.

Auditeur is a general-purpose, energy-efficient, and context-aware acoustic event detection platform for smartphones. It enables app developers to have their app register for and get notified on a wide variety of acoustic events. Auditeur is backed by a cloud service to store user contributed sound clips and to generate an energy-efficient and context-aware classification plan for the phone. When an acoustic event type has been registered, the smartphone instantiates the necessary acoustic processing modules and wires them together to execute the plan. The phone then captures, processes, and classifies acoustic events locally and efficiently. Our analysis on user-contributed empirical data shows that Auditeur's energy-aware acoustic feature selection algorithm is capable of increasing the device lifetime by 33.4%, sacrificing less than 2% of the maximum achievable accuracy. We implement seven apps with Auditeur, and deploy them in real-world scenarios to demonstrate that Auditeur is versatile, 11.04% - 441.42% less power hungry, and 10.71% - 13.86% more accurate in detecting acoustic events, compared to state-of-the-art techniques. We present a user study to demonstrate that novice programmers can implement the core logic of interesting apps with Auditeur in less than 30 minutes, using only 15 - 20 lines of Java code.

Mobile phones have become an ideal platform for physiological and environmental sensing. A number of research and commercial smartphone "accessories" have emerged in recent years that try to extend the sensing capabilities of a mobile phone. However, the major drawback of these devices is that they either require the user to act in some specific way or change their lifestyle and habit to some extent. In this demo, we present Septimu V2 (Septimu 2 ) -- a novel non-intrusive physiological and environmental sensing platform which is fully embedded in a conventional earphone, works with existing smartphones, and does not require the user to change habits in any way. Septimu 2 is a continuation of [1], and integrates a suite of new sensors. In addition to 3-axis accelerometer and gyroscope, Septimu 2 incorporates remote IR temperature sensor, IR LED, IR photodiode and two additional microphones. The baseboard performs signal condition and sends the data to cellphone via Bluetooth. Septimu 2 enables a number of applications, including heart-rate monitoring, fine grained posture detection, and external sound source localization and classification.

A mobile phone, as a pervasive device, has great potential in human wellness monitoring. In this demo, we first present the design and implementation of our hardware - SEPTIMU. SEPTIMU consists of a small baseboard and a pair of tiny sensor boards embedded inside conventional earphones. The baseboard provides power conversion and data communication through the normal audio jack interface. The embedded sensor board is 1×1cm2 and integrates 3-axis accelerometer, gyroscope, thermometer, photodiode and microphone. Secondly, we evaluate SEPTIMU using a mobile application that continuously monitors body posture and provides feedback to the user.

MusicalHeart is a biofeedback-based, context-aware, automated music recommendation system for smartphones. We introduce a new wearable sensing platform, Septimu, which consists of a pair of sensor-equipped earphones that communicate to the smartphone via the audio jack. The Septimu platform enables the MusicalHeart application to continuously monitor the heart rate and activity level of the user while listening to music. The physiological information and contextual information are then sent to a remote server, which provides dynamic music suggestions to help the user maintain a target heart rate. We provide empirical evidence that the measured heart rate is 75% -- 85% correlated to the ground truth with an average error of 7.5 BPM. The accuracy of the person-specific, 3-class activity level detector is on average 96.8%, where these activity levels are separated based on their differing impacts on heart rate. We demonstrate the practicality of MusicalHeart by deploying it in two real world scenarios and show that MusicalHeart helps the user achieve a desired heart rate intensity with an average error of less than 12.2%, and its quality of recommendation improves over time.

One of the most commonly identied precipitant of seizures for people with epilepsy is stress. We have developed a system for tracking objective measures in the subjects homes. An integrated system is used for data collection, uploading, and viewing the data. Our system is currently deployed and actively collecting data from people suering from epilepsy