Shuangquan Wang

Extreme learning machine-based device displacement free activity recognition model

Activity recognition based on mobile device is an important aspect in developing human-centric pervasive applications like gaming, industrial maintenance and health monitoring. However, the data distribution of accelerometer is heavily affected by varying device locations and orientations, which will degrade the performance of recognition model. To solve this problem, we propose a fast, robust and device displacement free activity recognition model in this paper, which integrates principal component analysis (PCA) and extreme learning machine (ELM) to realize location-adaptive activity recognition. On the one hand, PCA is employed to reduce the dimensionality of feature space and extract robust features for recognition. On the other hand, in the online phase ELM is applied to classify the activity and adapt the recognition model to new device locations based on high confident recognition results in real time. Experimental results show that, with robust features and fast adaptation capability, the proposed model can adapt the classifier to new device locations quickly and obtain good recognition performance.

FallAlarm: Smart Phone Based Fall Detecting and Positioning System

Abstract This paper presents a system called FallAlarm, which is utilized for fall detecting and positioning using a smart phone. A tri-axial accelerometer sensor and a Wi-Fi module embeded in the phone are employed to provide needed information. Data from the accelerometer is evaluated with a decision tree model to determine a fall. If a fall is suspected, a notification is raised to require the users response. If the user is injured hardly and cannot respond in time, the system immediately begin to position the occurence of the fall event by detected surrounding Wi-Fi signals, then automatically send a alarm message to his pre-specified guardian with a message via SMS(Short Message System). Consequently, the victim of fall can be monitored and cared in real time. Tested on a real-world data set, the FallAlarm system can achieve an acceptable accuracy for practical application.

A Class Incremental Extreme Learning Machine for Activity Recognition

Automatic activity recognition is an important problem in cognitive systems. Mobile phone-based activity recognition is an attractive research topic because it is unobtrusive. There are many activity recognition models that can infer a user’s activity from sensor data. However, most of them lack class incremental learning abilities. That is, the trained models can only recognize activities that were included in the training phase, and new activities cannot be added in a follow-up phase. We propose a class incremental extreme learning machine (CIELM). It (1) builds an activity recognition model from labeled samples using an extreme learning machine algorithm without iterations; (2) adds new output nodes that correspond to new activities; and (3) only requires labeled samples of new activities and not previously used training data. We have tested the method using activity data. Our results demonstrated that the CIELM algorithm is stable and can achieve a similar recognition accuracy to the batch learning method.

Inferring social contextual behavior from bluetooth traces

Context-aware computing is increasingly paid much attention, especially makes the peoples social contextual behavior very crucial for user-centric dynamic behavior inference. At present, extensive work has focused on detecting specific places inferred by static radio signals like GPS, GSM and WiFi, and recognizing mobility modes inferred by embedded sensor components like accelerometer. This paper proposes a distinct feature based classification approach and context restraint based majority vote rule to infer social contextual behavior in dynamic surroundings. Experimental results indicate that our proposed method can achieve high accuracy for inferring social contextual behavior through the real-life Bluetooth traces.

b-COELM

Various mini-wearable devices have emerged in the past few years to recognize activities of daily living for users. Wearable devices are normally designed to be miniature and portable. Models running on the devices inevitably face following challenges: low-computational-complexity, lightweight and high-accuracy. In order to meet these requirements, a novel powerful activity recognition model named b-COELM is proposed in this paper. b-COELM retains the superiorities (low-computational-complexity, lightweight) of Proximal Support Vector Machine, and extends the powerful generalization ability of Extreme Learning Machine in multi-class classification problems. Experimental results show the efficiency and effectiveness of b-COELM for recognizing activities of daily living.

Online sequential ELM based transfer learning for transportation mode recognition

Transportation mode recognition plays an important role in discovering life patterns from peoples physical behavior. Learning knowledge from mobile sensing data enables transportation mode recognition on mobile phone. However, existing transportation mode recognition methods are mostly based on fixed recognition models, which do not consider the diversities in different users and their transportation context. In this paper, an online sequential extreme learning machine based transfer learning method (TransELM) is proposed to recognize various transportation modes. TransELM is mainly comprised of three steps: firstly, an initial ELM classifier is trained on the labeled training data from the source domain; secondly, the mean and standard deviation are calculated as multi-class trustable intervals in source domain, and then the partially trustable samples are effectively extracted from the target domain; thirdly, the trustable samples are integrated, where an incremental OSELM method is employed to update the original ELM classifier. Experimental results show that TransELM obtains higher accuracy than the traditional ELM classifier in real world transportation mode recognition problems.

RECOGNIZING TRANSPORTATION MODE ON MOBILE PHONE USING PROBABILITY FUSION OF EXTREME LEARNING MACHINES

As one important clue to understand peoples behavior and life pattern, transportation mode (such as walking, bicycling, taking bus, driving, taking light-rail or subway, etc.) information has already widely used in mobile recommendation, route planning, social networking and health caring. This paper proposes a transportation mode recognition method using probability fusion of extreme learning machines (ELMs). Two ELM classification models are trained to recognize accelerometer data and Global Positioning System (GPS) data, respectively. Fuzzy output vectors of these two ELMs are transformed into probability vectors and fused to determine the final result. Experimental results verify that the proposed method is effective and can obtain higher recognition accuracy than traditional fusion methods.

PPCare: A Personal and Pervasive Health Care System for the Elderly

Over the past decade, the ageing of population has increasingly become a serious social issue all over the world. Accordingly, the market demand for health care monitoring of the elderly has greatly emerged. However, existing health care systems yet cannot provide convenient and comprehensive health care services for the elderly. This paper presents a mobile phone based personal and pervasive health care system for the elderly to monitor their daily life and physiological indexes. The whole system consists of four major functions, including sport planning, calories consumption estimation, fall alert and physiological indexes monitoring. PPCare system has been deployed and applied to the elderly in two communities for trial use and some constructive feedbacks have been collected for further system improvement.

A supervised learning based semantic location extraction method using mobile phone data

Various kinds of location-aware computing and applications are proliferating rapidly nowadays, which makes the location the most critical ingredient. However, on one hand, one location represented as the semantic meaning like “home” is more understandable than conveying the absolute physical coordinate; on the other hand, detected wireless data is a series of random sequence and the formed training vector has not equal-length feature, which may heavily leads to unstable accuracy of location extraction model because of varying human and environment factors. To robustly discover the users semantic locations in dynamic wireless environment, we propose a novel Hidden Markov Model (HMM)-based Location Extraction algorithm called HLE, which adopts a supervised learning based method for extracting users daily significant semantic locations using mobile phone data. We carry out the HLE algorithm on realistic wireless signal data, experimental results show that the proposed method is reasonable and effective for semantic location extraction in the real-world application.

Unobtrusive Sensing Incremental Social Contexts Using Class Incremental Learning

By utilizing captured characteristics of surrounding contexts through widely used Bluetooth sensor, user-centric social contexts can be effectively sensed and discovered by dynamic Bluetooth information. At present, state-of-the-art approaches for building classifiers can basically recognize limited classes trained in the learning phase; however, due to the complex diversity of social contextual behavior, the built classifier seldom deals with newly appeared contexts, which results in degrading the recognition performance greatly. To address this problem, we propose, an OSELM (online sequential extreme learning machine) based class incremental learning method for continuous and unobtrusive sensing new classes of social contexts from dynamic Bluetooth data alone. We integrate fuzzy clustering technique and OSELM to discover and recognize social contextual behaviors by real-world Bluetooth sensor data. Experimental results show that our method can automatically cope with incremental classes of social contexts that appear unpredictably in the real-world. Further, our proposed method have the effective recognition capability for both original known classes and newly appeared unknown classes, respectively.