• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.2
• /
• pp.13-20
• /
• 2020

One of the major issues surrounding big data is the availability of massive time-based or telemetry data. Now, the appearance of low cost capture and storage devices has become possible to get very detailed time data to be used for further analysis. Thus, we can use these time data to get more knowledge about the underlying system or to predict future events with higher accuracy. In particular, it is very important to define custom tailored contract offers for many households and businesses having smart meter records and predict the future electricity usage to protect the electricity companies from power shortage or power surplus. It is required to identify a few groups with common electricity behavior to make it worth the creation of customized contract offers. This study suggests big data transformation as a side effect and clustering technique to understand the electricity usage pattern by using the open data related to smart meter and KNIME which is an open source platform for data analytics, providing a user-friendly graphical workbench for the entire analysis process. While the big data components are not open source, they are also available for a trial if required. After importing, cleaning and transforming the smart meter big data, it is possible to interpret each meter data in terms of electricity usage behavior through a dynamic time warping method.

• Journal of Internet Computing and Services
• /
• v.14 no.6
• /
• pp.85-94
• /
• 2013

With the emergence of world wide web, in particular web 2.0 the rapidly growing amount of news articles has created a problem for users in selection of news articles according to their requirements. To overcome this problem different clustering mechanism has been proposed to broadly categorize news articles. However these techniques are totally machine oriented techniques and lack users' participation in the process of decision making for membership of clustering. In order to overcome the issue of zero-participation in the process of clustering news articles in this paper we have proposed a framework for clustering news articles by combining users' judgments that they post on twitter with the news articles to cluster the objects. We have employed twitter hash-tags for this purpose. Furthermore we have computed the credibility of users' based on frequency of retweets for their tweets in order to enhance the accuracy of the clustering membership function. In order to test performance of proposed methodology, we performed experiments on tweets messages tweeted during general election 2013 in Pakistan. Our results proved over claim that using users' output better outcome can be achieved then ordinary clustering algorithms.

• Journal of the Korean Geosynthetics Society
• /
• v.10 no.4
• /
• pp.61-70
• /
• 2011

This paper describes prediction results of penetration rate using case history in order to compare empirical models for penetration rate prediction of TBM. The reasonable empirical model is evaluated by comparison with prediction results and measured result. The penetration rate prediction is applied in separate empirical models considering rock characteristics and mechanical characteristics of TBM. The rock of applied filed had almost gneiss and its unconfined compressive strength was irregular due to the exist of weak zones and joint. In prediction results using unconfined compressive strength, Graham's model (1976) had impractical result when it had lower strength. NTNU model (1998) of the separate empirical models used in average penetration rate had the highest accuracy by comparison with the others, because it is a reasonable model which has rock characteristics and mechanical characteristics of TBM. However, Tarkoy's model (1986) based on unconfined compressive strength correspond with the measured values in field. Therefore, it should be considered a rock type, geological characteristic and mechanical characteristic of TBM at prediction of penetration rate.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.12
• /
• pp.5019-5024
• /
• 2015

Background: The purpose of this study was to assess the dosimetric and clinical feasibility of volumetric modulated arc based hypofractionated stereotactic radiotherapy (RapidArc) treatment for large acoustic schwannoma (AS >10cc). Materials and Methods: Ten AS patients were immobilized using BrainLab mask. They were subject to multimodality imaging (magnetic resonance and computed tomography) to contour target and organs at risk (brainstem and cochlea). Volumetric modulated arc therapy (VMAT) based stereotactic plans were optimized in Eclipse (V11) treatment planning system (TPS) using progressive resolution optimizer-III and final dose calculations were performed using analytical anisotropic algorithm with 1.5 mm grid resolution. All AS presented in this study were treated with VMAT based HSRT to a total dose of 25Gy in 5 fractions (5fractions/week). VMAT plan contains 2-4 non-coplanar arcs. Treatment planning was performed to achieve at least 99% of PTV volume (D99) receives 100% of prescription dose (25Gy), while dose to OAR's were kept below the tolerance limits. Dose-volume histograms (DVH) were analyzed to assess plan quality. Treatments were delivered using upgraded 6 MV un-flattened photon beam (FFF) from Clinac-iX machine. Extensive pretreatment quality assurance measurements were carried out to report on quality of delivery. Point dosimetry was performed using three different detectors, which includes CC13 ion-chamber, Exradin A14 ion-chamber and Exradin W1 plastic scintillator detector (PSD) which have measuring volume of $0.13cm^3$, $0.009cm^3$ and $0.002cm^3$ respectively. Results: Average PTV volume of AS was 11.3cc (${\pm}4.8$), and located in eloquent areas. VMAT plans provided complete PTV coverage with average conformity index of 1.06 (${\pm}0.05$). OAR's dose were kept below tolerance limit recommend by American Association of Physicist in Medicine task group-101(brainstem $V_{0.5cc}$ < 23Gy, cochlea maximum < 25Gy and Optic pathway <25Gy). PSD resulted in superior dosimetric accuracy compared with other two detectors (p=0.021 for PSD.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.22 no.6
• /
• pp.623-641
• /
• 2020

About 5 km length of tunnels were constructed by mechanized tunnelling method using closed type shield TBM. In order to avoid construction delay problems for ensuring timely electricity transmission, it is necessary to increase the prediction accuracy of the excavation process involving machines according to rock mass types. This is important to corroborate the project duration and optimum operation for various considerations involved in the machine. So, full-scale tunnelling tests were performed for developing the advance rate model to be appropriately used for 3.6 m diameter shield TBM. About 100 test cases were established and performed using various operational parameters such as thrust force and rotational speed of cuttterhead in representative uniaxial compressive strengths. Accordingly, relationships between normal force and penetration depth and, between UCS and torque were suggested which consider UCS and thrust force conditions according to weathered, soft, hard rocks. Capacity analysis of cutterhead was performed and optimum operational conditions were also suggested based on the developed model. Based on this study, it can be expected that the project construction duration can be reduced and users can benefit from the provision of earlier service.

• KIISE Transactions on Computing Practices
• /
• v.23 no.1
• /
• pp.57-64
• /
• 2017

Short message service (SMS) is one of the most important communication methods for people who use mobile phones. However, illegal advertising spam messages exploit people because they can be used without the need for friend registration. Recently, spam message filtering systems that use machine learning have been developed, but they have some disadvantages such as requiring many calculations. In this paper, we implemented a spam message filtering system using the set-based POI search algorithm and sentence similarity without servers. This algorithm can judge whether the input query is a spam message or not using only letter composition without any server computing. Therefore, we can filter the spam message although the input text message has been intentionally modified. We added a specific preprocessing option which aims to enable spam filtering. Based on the experimental results, we observe that our spam message filtering system shows better performance than the original set-based POI search algorithm. We evaluate the proposed system through extensive simulation. According to the simulation results, the proposed system can filter the text message and show high accuracy performance against the text message which cannot be filtered by the 3 major telecom companies.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.4
• /
• pp.825-832
• /
• 2014

Previous works for detecting arrhythmia have mostly used nonlinear method such as artificial neural network, fuzzy theory, support vector machine to increase classification accuracy. Most methods require accurate detection of P-QRS-T point, higher computational cost and larger processing time. Even if some methods have the advantage in low complexity, but they generally suffer form low sensitivity. Also, it is difficult to detect PVC accurately because of the various QRS pattern by person's individual difference. Therefore it is necessary to design an efficient algorithm that classifies PVC based on QRS pattern in realtime and decreases computational cost by extracting minimal feature. In this paper, we propose PVC classification based on QRS pattern using QS interval and R wave amplitude. For this purpose, we detected R wave, RR interval, QRS pattern from noise-free ECG signal through the preprocessing method. Also, we classified PVC in realtime through QS interval and R wave amplitude. The performance of R wave detection, PVC classification is evaluated by using 9 record of MIT-BIH arrhythmia database that included over 30 PVC. The achieved scores indicate the average of 99.02% in R wave detection and the rate of 93.72% in PVC classification.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_3
• /
• pp.1077-1093
• /
• 2020

Sea Surface Temperature (SST) is an important environmental indicator that affects climate coupling systems around the world. In particular, coastal regions suffer from abnormal SST resulting in huge socio-economic damage. This study used Long Short Term Memory (LSTM) and Convolutional Long Short Term Memory (ConvLSTM) to predict SST up to 7 days in the south sea region in South Korea. The results showed that the ConvLSTM model outperformed the LSTM model, resulting in a root mean square error (RMSE) of 0.33℃ and a mean difference of -0.0098℃. Seasonal comparison also showed the superiority of ConvLSTM to LSTM for all seasons. However, in summer, the prediction accuracy for both models with all lead times dramatically decreased, resulting in RMSEs of 0.48℃ and 0.27℃ for LSTM and ConvLSTM, respectively. This study also examined the prediction of abnormally high SST based on three ocean heatwave categories (i.e., warning, caution, and attention) with the lead time from one to seven days for an ocean heatwave case in summer 2017. ConvLSTM was able to successfully predict ocean heatwave five days in advance.

• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1535-1543
• /
• 2015

The brain-computer interface obtains a user's electroencephalogram as a replacement communication unit for the disabled such that the user is able to control machines by simply thinking instead of using hands or feet. In this paper, we propose a feature extraction method based on a non-selected filter by SBCSP to classify motor imagery EEG. First, we divide frequencies (4~40 Hz) into 4-Hz units and apply CSP to each Unit. Second, we obtain the FLD score vector by combining FLD results. Finally, the FLD score vector is projected onto the optimal plane for classification using PCA. We use BCI Competition III dataset IVa, and Extracted features are used as input for LS-SVM. The classification accuracy of the proposed method was evaluated using $10{\times}10$ fold cross-validation. For subjects 'aa', 'al', 'av', 'aw', and 'ay', results were $85.29{\pm}0.93%$, $95.43{\pm}0.57%$, $72.57{\pm}2.37%$, $91.82{\pm}1.38%$, and $93.50{\pm}0.69%$, respectively.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.5
• /
• pp.169-176
• /
• 2020

Recently, deep neural networks (DNNs) have brought revolutions to many mobile and embedded devices by providing human-level machine intelligence for various applications. However, high inference accuracy of such DNNs comes at high computational costs, and, hence, there have been significant efforts to reduce computational overheads of DNNs either by compressing off-the-shelf models or by designing a new small footprint DNN architecture tailored to resource constrained devices. One notable recent paradigm in designing small footprint DNN models is sharing parameters in several layers. However, in previous approaches, the parameter-sharing techniques have been applied to large deep networks, such as ResNet, that are known to have high redundancy. In this paper, we propose a parameter-sharing method for already parameter-efficient small networks such as ShuffleNetV2. In our approach, small templates are combined with small layer-specific parameters to generate weights. Our experiment results on ImageNet and CIFAR100 datasets show that our approach can reduce the size of parameters by 15%-35% of ShuffleNetV2 while achieving smaller drops in accuracies compared to previous parameter-sharing and pruning approaches. We further show that the proposed approach is efficient in terms of latency and energy consumption on modern embedded devices.