• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.5 no.4
• /
• pp.1-8
• /
• 1985

Stability analysis of the embankment as to water level varation is the most important problem in the safety of the slope because the stress of embankment inside varies as to drawdown of seepage line. Especially when the water level is rapidly drawdown, because the flow direction of the free surface changes the toe of embankment, the factor of safety comes to small, therefore the embankment is dangered. For the purpose of studing these phenomena, the experimental models are built with sand in the laboratory. In the experimental consideration, the falling seepage line and the shape of failure are measured. This paper intends to study the failure slip surface, the relationship between the factor of safety and drawdown velocity, and hydraulic gradient. The results of the experimental study are summarized as follows; 1. Owing to the drawdown of free surface, sliding failure occurred in the upstream fill, the height of failure is 5~10, 9~15, and 13~21(cm) in each model. 2. In consideration of the distribution of pore water pressure Table-5 shows each factor of safety. In the relationship between the drawdown velocity and the factor of factor it's velocity should be limited to 0.21~0.28 (cm/sec), according to each models. In the relationship between the factor of safety and the hydraulic gradient within the upstream slope, it's gradient must be below 0.36~0.43.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.28 no.2
• /
• pp.73-80
• /
• 2016

To reduce the mechanical energy loss and to get the high energy efficiency, an apparatus of wave power generation inducing a consistent one way rotating motion from the wave reciprocation motions was developed and the hydraulic experiments for the real electric power production were conducted and the results were discussed. In the experiments for the shape of the buoyant tank, the efficiency of the fixed 9 cm diameter type enduring the wave plate weight was 14.6% and this was the best result for all shapes. But although the free sliding type was expected to represent a high efficiency, the experiments did not show a good result as 8.5% efficiency. Therefore, the shape of buoyant tank was decided as the fixed 9 cm diameter type in the next all tests. In the experiments for the various incident waves, when the water depth was 90 cm, the average efficiencies were measured as 3.9% in the 2nd gear, 4.9% in the 3rd gear, 4.9% in the 4th gear, 12.0% in the 5th gear, 10.0% in the 6th gear, 3.1% in the 7th gear, and 3.0% in the 8th gear. Also, when the water depth was 80 cm, the average efficiency was shown as 15.0% with 5th gear condition. Therefore the high average efficiency as 13.5% was given with 80~90 cm water depth and the 5th gear in the model.

• Journal of the Korean Society for Precision Engineering
• /
• v.26 no.2
• /
• pp.101-108
• /
• 2009

A fretting wear test rig for dry ambient condition, which employs a piezoelectric actuator, has been developed. It is driven and loaded in a very simple manner with acceptable experimental accuracy. By using the rig, Inconel 690 tube has been tested under the normal load of 10 and 15N with sliding amplitude of less than $100{\mu}m$ during $10^6$cycles. The wear resistance of the material has been characterized in terms of the wear coefficient based on the work rate model. SEM micrographs show the complex structures of the scars, which consist of risen peaks, plate-type thin layers and locally exposed bare surfaces. The cracks spread over the layers give clue to the fretting wear mechanism of the material.

• The korean journal of orthodontics
• /
• v.37 no.2 s.121
• /
• pp.137-149
• /
• 2007

Objective: The purpose of this study was to compare changes in frictional resistance between the bracket and wire under dry and wet conditions according to a change in moment. Methods: A stainless steel bracket of $0.022"{\times}0.028"$ slot, and $0.019"{\times}0.025"$ stainless steel, beta-titanium, and nickel-titanium wires were used. A 10 mm length lever was attached to the test (sliding) brackets to generate a moment. The experimental model was designed to allow tipping until contacts were established between the wire and the mesiodistal edges of the bracket slot. The moment was generated by suspending a 100 g or 200 g weight on the end of the lever. The moments applied were $1000g{\cdot}mm\;(100g{\times}10mm)\;and\;2000g{\cdot}mm\;(200g{\times}10mm)$. The test brackets were ligated with elastomeric ligature for a constant ligation force and the fixed brackets were ligated with stainless steel ligature. Brackets were moved along the wire by means of an universal testing machine, and maximum frictional resistances were recorded. Results: Stainless steel wire showed least frictional resistance and there was no significant difference between beta-titanium and nickel-titanium except at $2000g{\cdot}mm$ moment in wet conditions. Frictional resistance of all wires increased as the moment increased from $1000g{\cdot}mm\;to\;2000g{\cdot}mm$. Under wet conditions, the frictional resistance of stainless steel wires increased in both $1000g{\cdot}mm\;and\;2000g{\cdot}mm$ moment conditions, but frictional resistance of nickel-titanium and beta-titanium increased only in $2000g{\cdot}mm$ conditions. Conclusion: These results indicated that various conditions influence on frictional resistance. Therefore, laboratory studies of frictional resistance should simulate clinical situation.

• Journal of the Korean Geotechnical Society
• /
• v.27 no.3
• /
• pp.41-54
• /
• 2011

The present study analyzes the stability of waterfront quay wall under the combined action of earthquake and tsunami. Adopting the limit equilibrium method, the stability of waterfront quay wall is checked for both the sliding and overturning. Forces due to tsunami are compared with the proposed formula and the 3-D one-field Model for immiscible TWO-Phase flows (TWOPM-3D). Variations of the stability of wall are also proposed by the parametric study including tsunami water height, horizontal seismic acceleration coefficient, internal friction angle of soil, friction angle between the wall and the soil and the pore water pressure ratio. The present study about the stability of wall is also compared with the case when earthquake and tsunami are not considered. As a result, the result of numerical analysis about the tsunami force is similar to that of proposed formula. When earthquake and tsunami are simultaneously considered, the stability of wall in passive case significantly decreases and tsunami forces in active case are affected as a resistance force on the wall and so the stability of wall increases.

• Journal of the Korean Geotechnical Society
• /
• v.29 no.1
• /
• pp.23-27
• /
• 2013

The risks of debris flows caused by climate change have increased significantly around the world. Recently, landslide disaster prevention technology is more focused on the failure and post-failure dynamics to mitigate the hazards in flow-prone area. In particular, we should define the soil strength and flow characteristics to estimate the debris flow mobility in the mountainous regions in Korea. To do so, we selected known ancient landslides area: Inje, Pohang and Sangju debris flows. Firstly we measured physical and mechanical properties: liquidity index and undrained shear strength by fall cone penetrometer. From the test results, we found that there is a possible relationship between liquidity index and undrained shear strength, $C_{ur}=(1.2/I_L)^{3.3}$, in the selected areas, even though they were different in geological compositions. Assuming that the yield stress is equal to the undrained shear strength at the initiation of sliding, we examined the flow characteristics of weathered soils in Korea. When liquidity index is given as 1, 1.5 and 3.0, the debris flow motion of weathered soils is compared with that of mud-rich sediments, which are known as low-activity clays. At $I_L=1$, it seems that debris flow could reach approximately 250m after 5 minutes. As liquidity index increased from 1 to 3, the debris flow propagation of weathered soils is twice than that of low-activity clays. It may be due to the fact that soil masses mixed with the ambient water and then highly fragmented during flow, thereby leading to the high mobility. The results may help to predict the debris flow propagation and to develop disaster prevention technology at similar geological settings, especially for the weathered soils, in Korea.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.24 no.3
• /
• pp.179-188
• /
• 2012

Performance analyses of vertical breakwaters were conducted for fictitiously designed breakwaters for various water depths to analyze the influence of climate change on the structures. The performance-based design method considering sea level rise and wave height increase due to climate change was used for the performance analysis. One of the problems of the performance-based design method is the large calculation time of wave transformation. To overcome this problem, the SWAN model combined with artificial neural network was used. The significant wave height and principal wave direction at the breakwater site are quickly calculated by using a trained neural network with inputs of deepwater significant wave height and principal wave direction, and tidal level. In general, structural stability becomes low due to climate change impacts, but the trend of stability is different depending on water depth. Outside surf zone, the influence of wave height increase becomes more significant, while that of sea level rise becomes negligible, as water depth increases. Inside surf zone, the influence of both wave height increase and sea level rise diminishes as water depth decreases, but the influence of wave height increase is greater than that of sea level rise. Reinforcement and maintenance policies for vertical breakwaters should be established with consideration of these results.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.15 no.3
• /
• pp.88-98
• /
• 2011

Masonry structure is a style of building which has been widely applied as residential facilities of low and middle stories, commercial and public facilities etc. But it is possible to destroy by loss of adhesive strength or sliding when lateral forces, such as earthquake, occurs. This study proposes a seismic retrofit method for masonry structure and its seismic performance is demonstrated by shaking table test. Two specimens per each shaking direction were made, having out-of-plane(weak axis) and in-plane(strong axis) direction. External load of 1 ton was also applied for each specimen during the test, to model the behavior of reinforced masonry wall. As a result of shaking table tests, it is shown that the specimen applying the proposed seismic retrofit method showed acceptable behaviors in both of Korea building design criteria(0.14g) and USA seismic criteria suggested by IBC(0.4g). However, it was observed that stiffness of the specimen toward out-of-plane was rapidly decreasing when seismic excitations over 0.14g were loaded. In comparison of relative displacements, maximum relative displacement of specimens which were accelerated toward out-of-plane with 0.4g at once was 29~31% of maximum relative displacement when specimens were gradually accelerated from 0.08g to 0.4g, while the maximum relative displacement of specimens accelerated toward in-plane has similar value in both cases. Therefore, it is concluded that the wall accelerated toward out-of-plane is more affected by hair crack or possible fatigues caused by seismic excitation.

• Management & Information Systems Review
• /
• v.33 no.1
• /
• pp.45-63
• /
• 2014

A trading system is a computer trading program that automatically submits trades to an exchange. Mechanical a trading system to execute trade is spreading in the stock market. However, a trading system to trade a single asset might occur instability of the profit because payoff of this system is determined a asset movement. Therefore, it is necessary to develop a trading system that is trade two assets such as a pair trading that is to sell overvalued assets and buy the undervalued ones. The aim of this study is to propose a relative value based trading system designed to yield stable and profitable profits regardless of market conditions. In fact, we propose a procedure for building a trading system that is based on the rough set analysis of indicators derived from a price ratio between two assets. KOSPI 200 index futures and S&P 500 index futures are used as a data for evaluation of the proposed trading system. We intend to examine the usefulness of this model through an empirical study.