研究者詳細 - 井上　創造

論文 - 井上　創造

分割表示226 件中 61 - 80 件目／全件表示 >>

Sensor-based Daily Activity Understanding in Caregiving Center 査読有り

Tahera Hossain, 井上創造

IEEE PerCom Student Forum 2 pages 2019年03月

IoT行動認識技術の実用化とオープンイノベーションへの取り組み

井上創造, 房前悟, 相馬功, 柴田智広

計測自動制御学会誌 ( 計測自動制御学会誌 ) 2019年02月

A Hybrid Model Using Hidden Markov Chain and Logic Model for Daily Living Activity Recognition 査読有り

Paula Lago, 井上創造

International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI 2018) 4 - 7 2018年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Punta Cana, Dominican Republic Punta Cana, Dominican Republic

We detail the solution to the UCAmI Cup Challenge to recognizing on going activities at home from sensor measurements. We use binary sensors and proximity sensor measurements for the recognition. We use an hybrid strategy, combining a probabilistic model and a definition-based model. The former consists of a Hidden Markov Model using the result of a neural network as emission probabilities. It is trained with the labelled data provided by the Cup. The latter approach takes advantage of the descriptions provided for each of the activities which are expressed in logical statements based on the sensors states. We then combine the results with a weighted average. We compare the performance of each individual strategy and of the combined strategy.

Kyutacar

その他リンク： https://www.mdpi.com/2504-3900/2/19/1266

IoT技術の医療介護への応用

井上創造

日医雑誌 ( 日医雑誌 ) 147 ( 8 ) 1619 - 1622 2018年11月

Activity Recognition: Translation Across Sensor Modalities Using Deep Learning

大北剛, 井上創造

Ubicomp Workshop on Human Activity Sensing Corpus and Applications (HASCA) 1462 - 1471 2018年10月

Supervised and Neural Classifiers for Locomotion Analysis 査読有り

Swapnil Sayan Saha, Shafizur Rahman, Tahera Hossain, 井上創造, Md Atiqur Rahman Ahad

Ubicomp Workshop on Human Activity Sensing Corpus and Applications (HASCA) 1563 - 1570 2018年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Singapore Singapore

The goal of the Sussex-Huawei Locomotion-Transportation (SHL) recognition challenge is to classify 8 modes of transportation and locomotion activities recorded using a smartphone inertial sensor. In this paper, Team Orion extracts 36 quantitative features per sensor (totalling 226 features) of the dataset provided in the SHL recognition challenge and provides a processing pipeline for training the classifiers embracing parallel computation and out-of-memory processing. One-vs-one quadratic Support Vector Machine (1-1 QSVM) and Bagged Decision Trees with 45 learners (EoC-45) were used for classification of the DU Mobility Dataset. The same features/pipeline provided 91% and 92% classification accuracies respectively on the SHL recognition challenge dataset. Using one-vs-one cubic Support Vector Machine (1-1 CSVM) on SHL, we received the highest classification accuracy of 92.8%. Afterwards, Artificial Neural Network (ANN) was applied on the dataset and accuracies of 93.6%, 88.18% and 90.05% were obtained for training, testing and validation phases. The results provide promising prospects of supervised and neural network classification for locomotion analysis.

その他リンク： https://dl.acm.org/citation.cfm?id=3267524

Study of LoRaWAN Technology for Activity Recognition 査読有り

Tahera Hossain, 土井祐介, Tahia Tazin, Md Atiqur Rahman Ahad, 井上創造

Ubicomp Workshop on Human Activity Sensing Corpus and Applications (HASCA) 5 pages 2018年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Singapore Singapore

In this paper, we explore LoRaWAN (Long Range Wide
Area Network) sensor for human activity recognition. In
this research, we want to verify activity recognition by
using LoRaWAN devices. We want to explore relation
between packet loss and activity recognition accuracy
from LoRaWAN sensor data. Afterwards, we study
LoRaWAN sensors data transfer ability in a real
caregiving center. We want to estimate the packet loss
amount from realistic sensors. In LoRaWAN technology,
the amount of sensor nodes connected with a single
gateway have an impact on the performance of sensors
ultimate data sending capability in terms of packet loss.
By exploring a single gateway, we transfer the
LoRaWAN sensor data to the cloud platform
successfully. We evaluate LoRaWAN accelerometer
sensors data for human activity recognition. We explore
the Linear Discriminant Analysis (LDA), Random Forest
(RnF) and K-Nearest Neighbor (KNN) for classification.
We achieve recognition accuracy of 94.44% by LDA,
84.72% by RnF and 98.61% by KNN. Then we simulate
the packet loss environment in our dataset to explore
the relation between packet loss and accuracy. In real
caregiving center, we did experiment with 42 LoRaWAN
sensors node connectivity and data transfer ability to
evaluate the packet received and packet loss
performance with LoRaWAN sensors.

その他リンク： http://hasca2018.hasc.jp/

MEASURed: A Multi-Sensor Setting Activity Recognition Simulation Tool 査読有り

武田紳吾, Paula Lago, 大北剛, 井上創造

Ubicomp Workshop on Human Activity Sensing Corpus and Applications (HASCA) 5 pages 2018年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Singapore Singapore

Motion capture generates data which are often more accurate
than those captured by multiple of accelerometer
sensors by their physical specification. Based on the observation
that accelerometer data can be virtually build by
the second derivation of position data from motion capture,
we propose a simulator, called MEASURed, for activity
recognition classifier. MEASURed can accommodate
any number of virtual accelerometer sensors on the
body based on some given motion capture data. Therefore,
MEASURed can evaluate activity recognition classifiers in
settings with different number, placement, and sampling
rate of accelerometer sensors. Our results show that the
F1-Score estimated by MEASURed is close to that obtained
with the real accelerometer data.

その他リンク： http://hasca2018.hasc.jp/

FonLog - A Mobile App for Nursing Activity Recognition 査読有り

Nattaya Mairittha (Fah), Tittaya Marittha (Fon), 井上創造

ACM Int'l Conf. Pervasive and Ubiquitous Computing (Ubicomp) Demo 4 pages 2018年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Singapore Singapore

We propose FonLog: a mobile app that can be used for a
data collection tool in human activity recognition for nursing
services. The app has essential features to collect activity
data efficiently found by the feedbacks from nursing facilities
from the staffs, namely, recording activity targets, a
user-friendly interface, recognition feedback, customizable
detailed records, instant activity, and offline functionality.

その他リンク： http://ubicomp.org/ubicomp2018/

Sensing Experiment in a Caregiving Facility for Correlation Analysis of Sleep and Daytime Activities 査読有り

松木萌, 井上創造

ACM Int'l Conf. Pervasive and Ubiquitous Computing (Ubicomp) Poster 4 pages 2018年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Singapore Singapore

We present that we get experiment for 5 months with 30 resident of a caregiving facility to collect big data and toanalyze of the correlation between sleep and daytime behavior. Analysis of sleep is useful for health care and selfanalysis. However, the exist sleep analysis application does not analyze correlation between sleep and the daytime activities such as what kind of activity affect from sleep, and what kind of activity should be taken to improve sleep quality. That is analysis of correlation between sleep and activities is important but it icollectins difficult to collect both the daytime activities data like "meal" and "toilet" and sleep data over a long period of time . In this paper, we get sensing experiment at the caregiving facility to collect big data for 30 subjects and for 5 months. Furthermore, by collecting care records for subject, it is possible to collect the day-
time activities data. Finally, we analyze the data of five subjects for 26 days, it was found that (1) the sleeping situation can estimate whether users will do daytime exercise is performed with an accuracy of 91% at the maximum (2) It was found that the information of the daytime exercise influences the time which user start of sleep.

その他リンク： http://ubicomp.org/ubicomp2018/

Activity Recognition by Using LoRaWAN Sensor 査読有り

Tahera Hossain, Tahia Tazin, Md Atiqur Rahman Ahad, 井上創造

ACM Int'l Conf. Pervasive and Ubiquitous Computing (Ubicomp) Poster 4 pages 2018年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Singapore Singapore

Low Power Wide Area (LPWA) technologies are
monumental for the IoT sector. Among the LPWA
technologies, long range (LoRa) and narrowband (NB)-
IoT are two promising technologies. In this paper, we
explore LoRaWAN (LoRa Wide Area Network) sensor for
human activity recognition. To support the ageing

population in the world, assisted living and sensor-
based human activity recognition are important

research areas. In this research, we propose an activity
recognition framework by exploiting LoRaWAN sensor
and its accelerometer data. In our framework, we
explore Arduino Uno, Arduino Lucky Shield having a
number of different sensors, and LoRaWAN to build one
compact system. By exploring a LoRaWAN Gateway, we
transfer the sensor data to SORACOM cloud platform
successfully. Then from the cloud data, a few statistical
features are computed to classify three activities such
as walk, stay and run. We aggregate the time series
data into different action labels that summarize the
user activity over a time interval. After, we train data
to induce a predictive model for activity recognition. We
explore the K-Nearest Neighbor (KNN) and Linear
Discriminant Analysis (LDA) for classification. We
achieve recognition accuracy 80% by KNN and 73.3%
by LDA. The result provides promising prospect for
LoRaWAN sensor for improving healthcare monitoring
service.

その他リンク： http://ubicomp.org/ubicomp2018/

Improving Sensor-based Activity Recognition Using Motion Capture as Additional Information 査読有り

Paula Lago, 武田紳吾, 大北剛, 井上創造

ACM Int'l Conf. Pervasive and Ubiquitous Computing (Ubicomp) Poster 4 pages 2018年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Singapore Singapore

We propose a new method for human activity recognition
using a single accelerometer sensor and additional sensors
for training. The performance of inertial sensors for complex

activities drops considerably compared with simple activ-
ities due to inter-class similarities. In such cases deploy-
ing more sensors may improve the performance. But such

strategy is often not feasible in reality due to costs or pri-
vacy concerns among others. In this context, we propose

a new method to use additional sensors only in training
phase. We introduce the idea of mapping the test data to
a codebook created from the additional sensor information.
Using the Berkeley MHAD dataset our preliminary results
show this worked positively; improving by 10.0% both the
average F1-score and the average accuracy. Notably, we
achieved an average improvement of 143.1% for the stand,
sit and sit to stand activities, typical activities for which the
inertial sensor is less informative when using the wrist-worn
accelerometer.

その他リンク： http://ubicomp.org/ubicomp2018/

Pre-Consulting Dialogue Systems for Telemedicine: Yes/No Intent Classification 査読有り

Tittaya Marittha (Fon), 大北剛, 井上創造

Ubicomp Workshop on Computing For Well-being (WellComp) 4 pages 2018年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Singapore Singapore

Symptoms estimation is important in the diagnosis of diseases. Pay attention to methods of assessing symptoms and collecting data on multiple symptoms that are keys to significantly enhance the effectiveness of the diagnosis, in addition, to make more informed decisions for both patients and doctors.
In this paper, we deal with symptoms estimation using the Twitter data, focusing on a diabetic retinopathy disease, to denote each answer as a "Yes" answer or a "No" answer. We propose a Long Short-Term Memory (LSTM) to show that the proposed method is effective.

その他リンク： http://wellcomp.org

介護施設紹介コールセンター記録のアンサンブル学習による将来予測と傾向分析査読有り

松木萌, 井上創造, 清田陽司

情報処理学会論文誌 2018年07月

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）

介護施設紹介コールセンターの記録データをアンサンブル学習を用いて傾向分析や将来予測を行い，「コールセンター」「介護施設」「相談者と入居者」の三者にとって有用な知見を取得する．介護分野における問題に対して，現場のデータを分析し問題解決に有用な知見を得る研究は多くあるが，それらは入居後の問題解決に向けたデータ分析がほとんどである．本研究では，介護施設入居前の分析も重要と考え，入居前の相談を受け付けるコールセンターの記録データを分析する．分析方法は，アンサンブル学習を用いて，説明変数の属性値を2値化し，属性の重要度と目的変数に対する正負の相関を同時に可視化した．記録データは，相談者1名に対し1行，「性別」や「介護度」のような属性を列とした，7,685行106列の行列データである．分析の結果，記録に欠損が多い人は成約に行かないことや，交通手段が見学に行くかどうかの重要な要因であること，「見学するかどうか」は機械学習による精度が96．8%であること，入居決定の決め手は「介護サービス」ではなく，「きれいさ」や「印象」，「距離」であることなど，10件ほどの知見を取得した．

Kyutacar

Virtual Realty Implementation of Steering Hands Motion in Driving Simulator with Unity Technology 査読有り

生駒哲一, 土井祐介, 井上創造

International Conference on Informatics, Electronics & Vision (ICIEV) & 2nd International Conference on Imaging, Vision & Pattern Recognition (icIVPR) 6 pages 2018年06月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Kitakyushu Kitakyushu

Abstract―Motivated to achieve visual feedback of body motion
in head mounted virtual reality, proposal of this paper is one
practical implementation method of a system that captures hands
motion of car driver in steering operation via vision sensor and
displays the driver’s hands over virtual reality graphics governed
by Unity technology. It utilizes TCP/IP socket communication
with client/server framework for management of passing data

among multiple computer as clients. Actual implementation re-
sult has been reported under driving simulator facility equipped

with two-axes motion actuators and HTC Vive head mounted
display for virtual reality based on Unity technology as graphics
and driving simulator software engine.

A Study on Sensor-based Activity Recognition Having Missing Data 査読有り

Tahera Hossain, 後藤広樹, Md Atiqur Rahman Ahad, 井上創造

International Conference on Informatics, Electronics & Vision (ICIEV) & 2nd International Conference on Imaging, Vision & Pattern Recognition (icIVPR) 6 pages 2018年06月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Kitakyushu Kitakyushu

Abstract― Human activity recognition is an important area for
various applications. Sensor-based activity recognition
deteriorates while partial data are lost. Hence, in this paper,
we study activity recognition in the presence of data loss.
Earlier, we explored sensor-based activity recognition where
we train the data with randomly missed data [1]. It is required
to investigate better features for handling missing data. Here,
we evaluate activity performance result with missing data
environment with various feature combinations for multiple
classifiers. Initially, we developed a simulated environment to
study the impact of features. Afterward, we evaluated our
proposed feature-based method on a benchmark dataset
named HASC dataset. The dataset has no missing data.
However, to evaluate our approach, we added various levels of
missing data randomly and studied the performances. We
explored mean, variance, skewness and kurtosis as statistical
features based on a time-windowing approach. For
classification study, we exploited two classifiers called Nave
Bayes and Random Forest. Our approach and study
demonstrated satisfactory recognition results under various
feature combinations in different situations of missing data.

Machine Learning of User Attentions in Sensor Data Visualization 査読有り

藤野慶汰, 井上創造, 柴田智広

EAI International Conference on Mobile Computing, Applications and Services (MobiCASE) 125 - 143 2018年03月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Osaka Osaka

In this paper, we propose a method to automatically esti- mate important points of large sensor data by collecting users attention points when visualized, and by applying supervised machine learning al- gorithm. For large scale sensor data, it is difficult to find important points just by visualization, because the points are buried in a large scope of visualization. We show the result of the estimation, where the accuracy was over 80% for multiple visualization. We also show the result of re- usability for new type of visualization, which performed still 70-80% of accuracies.

その他リンク： http://mobicase.org/2018/show/home

Dialogue Breakdown Detection with Long Short Term Memory 査読有り

Tittaya Marittha (Fon), 大北剛, 井上創造

EAI International Conference on Mobile Computing, Applications and Services (MobiCASE) poster 245 - 250 2018年03月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Osaka Osaka

This paper aims to detect the utterance which can be categorized as the breakdown of the dialogue flow. We propose a logistic regression-based and a Long Short-Term Memory (LSTM)-based methods. Using the input with utterance-response pairs the performance of the LSTM-based method is superior to that of the logistic regression-based method in 36% measured with F-measure. We also measured the performance using the performance with utterance-response pairs: the performance with the input only with responses is unexpectedly inferior to those with responses in 6% to 23% measured with F-measure.

その他リンク： http://mobicase.org/2018/show/home

Gamification for High-quality Dataset in Mobile Activity Recognition 査読有り

Nattaya Mairittha (Fah), 井上創造

EAI International Conference on Mobile Computing, Applications and Services (MobiCASE) poster 216 - 222 2018年03月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Osaka Osaka

This paper presents a gamification concept for getting high- quality user-annotated datasets in the context of mobile activity recogni- tion. The‘novel’idea behind this concept is that users are motivated by getting feedback about the quality of their labeling activity as rewards or gamification element. For that, the collected sensor data and labels are used as training data for a machine learning algorithm for determining the dataset quality based on the resulting accuracy. By using the pro- posed method, the results show that the gamification elements increase the quantity (the proposed method is greater than the naive by at least 305) and‘ quality ’(the accuracy of the proposed data outperformed the original data by at least 4.3%) of the labels. Besides, the cheating de- tection algorithm could detect cheating with the accuracy of more than 70% that is fascinating work.

その他リンク： http://mobicase.org/2018/show/home

Improving Activity Recognition for Missing Data 査読有り

Tahera Hossain, 後藤広樹, 井上創造

EAI International Conference on Mobile Computing, Applications and Services (MobiCASE) Student Workshop 4 pages 2018年02月

　前のページ - 次のページ　

<井上 創造>

論文 - 井上 創造

<井上　創造>

論文 - 井上　創造