Browsing by Author "Njoo, Gunarto Sindoro"

Now showing 1 - 6 of 6

A Fusion-based Approach for User Activities Recognition on Smart Phones
(2015)
Njoo, Gunarto Sindoro
;
Ruan, Xiao Wen
;
Hsu, Kuo-Wei
;
Peng, Wen-Chih
;
交大名義發表
In the recent years, several research works have been conducted on collecting context data from various sensors for activity inference. We observe that users perform several actions in their mobile phones: taking photos, performing checkins, and accessing Wi-Fi networks. These actions generate spatial-temporal data that could be utilized to capture user activities. Spatial-temporal data could indicate that a user stays in a certain location at a particular time for a certain activity. In addition, by referring to social media data, one could also infer user activities. Three types of features are extracted for activity inference: 1) geographical feature, indicating where a user performs activities; 2) temporal feature, indicating when a user performs activities; and 3) semantic feature, showing the semantic concept of a place from location-based social networks. Here, we propose Spatial-Temporal Activity Inference Model (STAIM) to infer user activities from data with those three features. In addition, to determine the weight for each feature, we further propose three methods based on frequency, entropy, and entropy-frequency. Experimental results show that STAIM is able to effectively infer user activities, achieving 75% accuracy on average. Moreover, STAIM could infer user activities even when there is no training data (with some performance loss). Moreover, sensitive analysis of parameters is also conducted to select the most optimal parameter.
Associative Classification for Human Activity Inference on Smart Phones
(2014-01-01)
Peng, Yu-Hsiang
;
Njoo, Gunarto Sindoro
;
Li, Shou-Chun
;
Peng, Wen-Chih
;
資訊工程學系
With the population of smart phones, the general trend of human activity inference is prospering under a powerful computation capabilities on modern phones. Such an assistant make users life more convenient and help them prevent from unnecessary interferences. In conventional research, the activity inference problem is considered a classification instance, so in this paper we propose an association-based classifier framework (ACF) that aims at exploring the correlation among collected sensor data. Each data consists of multiple sensor readings with a label, e.g., dining, shopping, working, driving, sporting, and entertaining. Note that ACF caters to the discrete data; as a consequence, the continuous sensor readings are needed to be transformed to some discrete groups. Therefore, we propose an Interval Length-Gini Discretization (LGD) method which considers the groups and misclassified cases to obtain the best hypothesis for a given set of data. After an appropriate discretization, we propose one-cut and memory-iteration-based approach to select a set of useful sensor-value pairs for reducing the model size by removing redundant features and guaranteeing an acceptable accuracy. In the experiments our framework has a good performance on real data set collected from 50 participants in eight months, and a smaller size than the existing classifications.
Distinguishing friends from strangers in location-based social networks using co-location
(2018-10-01)
Njoo, Gunarto Sindoro
;
Hsu, Kuo-Wei
;
Peng, Wen-Chih
;
交大名義發表
;
National Chiao Tung University
This paper particularly focuses on using the spatiotemporal data in the location-based social networks (LBSNs) to infer the social tie between two users. To do so, we first generate a co-location dataset by simulating the meeting event between users based on the time difference and spatial distance. Subsequently, we extract four key features from the generated dataset: diversity, popularity, duration, and stability. We propose a framework called SCI (Social Connection Inference) that integrates all derived features to distinguish real friends' meetings from strangers' coincidental meetings. Experiment results based on the three LBSN datasets prove the effectiveness of the proposed SCI framework by outperforming the state-of-the-art methods. In addition, various discussions on different aspects of the data are presented in this paper to yield insights into using generated co-location datasets. (C) 2018 Elsevier B.V. All rights reserved.
Exploring Multi-View Learning for Activity Inferences on Smartphones
(2016-01-01)
Njoo, Gunarto Sindoro
;
Lai, Chien-Hsiang
;
Hsu, Kuo-Wei
;
交大名義發表
;
National Chiao Tung University
Inferring activities on smartphones is a challenging task. Prior works have elaborated on using sensory data from built-in hardware sensors in smartphones or taking advantage of location information to understand human activities. In this paper, we explore two types of data on smartphones to conduct activity inference: 1) Spatial-Temporal: reflecting daily routines from the combination of spatial and temporal patterns, 2) Application: perceiving specialized apps that assist the user's activities. We employ multi-view learning model to accommodate both types of data and use weighted linear kernel model to aggregate the views. Note that since resources of smartphones are limited, activity inference on smartphones should consider the constraints of resources, such as the storage, energy consumption, and computation power. Finally, we compare our proposed method with several classification methods on a real dataset to evaluate the effectiveness and performance of our method. The experimental results show that our approach outperforms other methods regarding the balance between accuracy, running time, and storage efficiency.
On Selecting Feature-Value Pairs on Smart Phones for Activity Inferences
(2014)
Njoo, Gunarto Sindoro
;
Peng, Yu-Hsiang
;
Hsu, Kuo-Wei
;
Peng, Wen-Chih
;
交大名義發表
Prior works have been elaborated on activity inferences from context information sensed on smart phones. Most of sensing computations are performed on CPU of smart phones. Thus, Sensor Hub is designed to avoid CPU involvement. However, Sensor Hub has several limitations, such as limited memory space and computation power. Since activity inference is a classification problem, prior works have already proposed some classifiers on smart phones. However, if one would like to build a classifier model in Sensor Hub, one challenging issue is to reduce the model size. One approach to reduce the model size of classifiers is feature selection. Feature selection reduces the model size by removing features in the feature set. Nonetheless, because sensor space is limited, removing features could reduce accuracy of classifier algorithm significantly. Therefore we explore feature-value selection concept, which considers the value rather than the feature to reduce the model size while preserving accuracy of classifiers. In this paper, we propose three feature-value selection methods, which consider confusion and redundancy among the feature-value. Due to the nature of feature-values, discretization of sensor data is important. We design a discretization method, LGD (Length Gini Discretization) and compare it with another method, MDLP (Minimum Description Length Discretization), to discretize sensor data using confusion metric to choose the cut point. Extensive experiments are conducted to evaluate our proposed feature-value selection methods. Feature-value removal allows us to reduce up to 80% of the model size and maintain average accuracy performance to 86%.
Understanding Human Behavior through Sensory Data and Location Based Services
(2019-01-01)
Njoo, Gunarto Sindoro
;
交大名義發表
;
National Chiao Tung University
Human behavior mining is an important task in the ubiquitous computing era. Delivering relevant services or information based on the users' activity context could bring great benefit to the users. The latest advances in smart devices (e.g., smartphones, smartwatches) unlock the possibility of understanding users' context continuously with a less obtrusive approach. However, unlike motion activities (e.g., sitting, standing, or walking), human activities (e.g., having a meeting, driving to the office, having lunch) require more sensory data which captures users' context at the moment and are more complex to be identified. Therefore, various information from multiple sources collaborates to model the users' behavior and pattern. In addition, due to space and energy limitation of smart devices, users' behavior model should be minified. To address this issue, we proposed a new preprocessing methods based on the selection of values in the data. Finally, even though human activity is highly personalized based on the users' preferences but we could not neglect the importance of social influence from close friends. Based on the mobility data generated, one could identify the social relationship between two people. All in all, this work is a collaboration work with Dr. Kuo-Wei Hsu(1) and Prof. Wen-Chih Peng(2).