• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.227-252
• /
• 2018

The evolution of instant communication has mirrored the development of the Internet and messenger applications are among the most representative manifestations of instant communication technologies. In messenger applications, senders use emoticons to supplement the emotions conveyed in the text of their messages. The fact that communication via messenger applications is not face-to-face makes it difficult for senders to communicate their emotions to message recipients. Emoticons have long been used as symbols that indicate the moods of speakers. However, at present, emoticon-use is evolving into a means of conveying the psychological states of consumers who want to express individual characteristics and personality quirks while communicating their emotions to others. The fact that companies like KakaoTalk, Line, Apple, etc. have begun conducting emoticon business and sales of related content are expected to gradually increase testifies to the significance of this phenomenon. Nevertheless, despite the development of emoticons themselves and the growth of the emoticon market, no suitable emoticon recommendation system has yet been developed. Even KakaoTalk, a messenger application that commands more than 90% of domestic market share in South Korea, just grouped in to popularity, most recent, or brief category. This means consumers face the inconvenience of constantly scrolling around to locate the emoticons they want. The creation of an emoticon recommendation system would improve consumer convenience and satisfaction and increase the sales revenue of companies the sell emoticons. To recommend appropriate emoticons, it is necessary to quantify the emotions that the consumer sees and emotions. Such quantification will enable us to analyze the characteristics and emotions felt by consumers who used similar emoticons, which, in turn, will facilitate our emoticon recommendations for consumers. One way to quantify emoticons use is metadata-ization. Metadata-ization is a means of structuring or organizing unstructured and semi-structured data to extract meaning. By structuring unstructured emoticon data through metadata-ization, we can easily classify emoticons based on the emotions consumers want to express. To determine emoticons' precise emotions, we had to consider sub-detail expressions-not only the seven common emotional adjectives but also the metaphorical expressions that appear only in South Korean proved by previous studies related to emotion focusing on the emoticon's characteristics. We therefore collected the sub-detail expressions of emotion based on the "Shape", "Color" and "Adumbration". Moreover, to design a highly accurate recommendation system, we considered both emotion-technical indexes and emoticon-emotional indexes. We then identified 14 features of emoticon-technical indexes and selected 36 emotional adjectives. The 36 emotional adjectives consisted of contrasting adjectives, which we reduced to 18, and we measured the 18 emotional adjectives using 40 emoticon sets randomly selected from the top-ranked emoticons in the KakaoTalk shop. We surveyed 277 consumers in their mid-twenties who had experience purchasing emoticons; we recruited them online and asked them to evaluate five different emoticon sets. After data acquisition, we conducted a factor analysis of emoticon-emotional factors. We extracted four factors that we named "Comic", Softness", "Modernity" and "Transparency". We analyzed both the relationship between indexes and consumer attitude and the relationship between emoticon-technical indexes and emoticon-emotional factors. Through this process, we confirmed that the emoticon-technical indexes did not directly affect consumer attitudes but had a mediating effect on consumer attitudes through emoticon-emotional factors. The results of the analysis revealed the mechanism consumers use to evaluate emoticons; the results also showed that consumers' emoticon-technical indexes affected emoticon-emotional factors and that the emoticon-emotional factors affected consumer satisfaction. We therefore designed the emoticon recommendation system using only four emoticon-emotional factors; we created a recommendation method to calculate the Euclidean distance from each factors' emotion. In an attempt to increase the accuracy of the emoticon recommendation system, we compared the emotional patterns of selected emoticons with the recommended emoticons. The emotional patterns corresponded in principle. We verified the emoticon recommendation system by testing prediction accuracy; the predictions were 81.02% accurate in the first result, 76.64% accurate in the second, and 81.63% accurate in the third. This study developed a methodology that can be used in various fields academically and practically. We expect that the novel emoticon recommendation system we designed will increase emoticon sales for companies who conduct business in this domain and make consumer experiences more convenient. In addition, this study served as an important first step in the development of an intelligent emoticon recommendation system. The emotional factors proposed in this study could be collected in an emotional library that could serve as an emotion index for evaluation when new emoticons are released. Moreover, by combining the accumulated emotional library with company sales data, sales information, and consumer data, companies could develop hybrid recommendation systems that would bolster convenience for consumers and serve as intellectual assets that companies could strategically deploy.

• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.127-148
• /
• 2020

The data center is a physical environment facility for accommodating computer systems and related components, and is an essential foundation technology for next-generation core industries such as big data, smart factories, wearables, and smart homes. In particular, with the growth of cloud computing, the proportional expansion of the data center infrastructure is inevitable. Monitoring the health of these data center facilities is a way to maintain and manage the system and prevent failure. If a failure occurs in some elements of the facility, it may affect not only the relevant equipment but also other connected equipment, and may cause enormous damage. In particular, IT facilities are irregular due to interdependence and it is difficult to know the cause. In the previous study predicting failure in data center, failure was predicted by looking at a single server as a single state without assuming that the devices were mixed. Therefore, in this study, data center failures were classified into failures occurring inside the server (Outage A) and failures occurring outside the server (Outage B), and focused on analyzing complex failures occurring within the server. Server external failures include power, cooling, user errors, etc. Since such failures can be prevented in the early stages of data center facility construction, various solutions are being developed. On the other hand, the cause of the failure occurring in the server is difficult to determine, and adequate prevention has not yet been achieved. In particular, this is the reason why server failures do not occur singularly, cause other server failures, or receive something that causes failures from other servers. In other words, while the existing studies assumed that it was a single server that did not affect the servers and analyzed the failure, in this study, the failure occurred on the assumption that it had an effect between servers. In order to define the complex failure situation in the data center, failure history data for each equipment existing in the data center was used. There are four major failures considered in this study: Network Node Down, Server Down, Windows Activation Services Down, and Database Management System Service Down. The failures that occur for each device are sorted in chronological order, and when a failure occurs in a specific equipment, if a failure occurs in a specific equipment within 5 minutes from the time of occurrence, it is defined that the failure occurs simultaneously. After configuring the sequence for the devices that have failed at the same time, 5 devices that frequently occur simultaneously within the configured sequence were selected, and the case where the selected devices failed at the same time was confirmed through visualization. Since the server resource information collected for failure analysis is in units of time series and has flow, we used Long Short-term Memory (LSTM), a deep learning algorithm that can predict the next state through the previous state. In addition, unlike a single server, the Hierarchical Attention Network deep learning model structure was used in consideration of the fact that the level of multiple failures for each server is different. This algorithm is a method of increasing the prediction accuracy by giving weight to the server as the impact on the failure increases. The study began with defining the type of failure and selecting the analysis target. In the first experiment, the same collected data was assumed as a single server state and a multiple server state, and compared and analyzed. The second experiment improved the prediction accuracy in the case of a complex server by optimizing each server threshold. In the first experiment, which assumed each of a single server and multiple servers, in the case of a single server, it was predicted that three of the five servers did not have a failure even though the actual failure occurred. However, assuming multiple servers, all five servers were predicted to have failed. As a result of the experiment, the hypothesis that there is an effect between servers is proven. As a result of this study, it was confirmed that the prediction performance was superior when the multiple servers were assumed than when the single server was assumed. In particular, applying the Hierarchical Attention Network algorithm, assuming that the effects of each server will be different, played a role in improving the analysis effect. In addition, by applying a different threshold for each server, the prediction accuracy could be improved. This study showed that failures that are difficult to determine the cause can be predicted through historical data, and a model that can predict failures occurring in servers in data centers is presented. It is expected that the occurrence of disability can be prevented in advance using the results of this study.

• Journal of Biomedical Engineering Research
• /
• v.6 no.2
• /
• pp.1-8
• /
• 1985

The wall shear stress in the vicinity of end-to end anastomoses under steady flow conditions was measured using a flush-mounted hot-film anemometer(FMHFA) probe. The experimental measurements were in good agreement with numerical results except in flow with low Reynolds numbers. The wall shear stress increased proximal to the anastomosis in flow from the Penrose tubing (simulating an artery) to the PTFE: graft. In flow from the PTFE graft to the Penrose tubing, low wall shear stress was observed distal to the anastomosis. Abnormal distributions of wall shear stress in the vicinity of the anastomosis, resulting from the compliance mismatch between the graft and the host artery, might be an important factor of ANFH formation and the graft failure. The present study suggests a correlation between regions of the low wall shear stress and the development of anastomotic neointimal fibrous hyperplasia(ANPH) in end-to-end anastomoses. 30523 T00401030523 ^x Air pressure decay(APD) rate and ultrafiltration rate(UFR) tests were performed on new and saline rinsed dialyzers as well as those roused in patients several times. C-DAK 4000 (Cordis Dow) and CF IS-11 (Baxter Travenol) reused dialyzers obtained from the dialysis clinic were used in the present study. The new dialyzers exhibited a relatively flat APD, whereas saline rinsed and reused dialyzers showed considerable amount of decay. C-DAH dialyzers had a larger APD(11.70$\pm$1.32mmHg/min)compared to CF dialyzers(4.32$\pm$0.55mmHg/min)(p<0.05). However, there was no observable difference in the UFR between the two dialyzers. Neither APD nor UFR showed any significant increase with an increasing number of reuses for up to more than 20reuses. A substantial number of failures observed in APD(larger than 20mmHe/min)on the reused dialyzers(2 out of 40 CP and S out 26 C-DAK) were attributed to the Possible damage on the fibers. The CF 15-11 HFDs which failed APD test did not show changes in the UFR compared to normal dialyzers indicating that APD is a more sensitive test than UFR test to evaluate the integrity of the fibers. 30527 T00401030527 ^x For quantitative measurement of reflected light from a clinical diagnostic strip, a prototype old reflectance photometer was designed. The strip loader and cassette were made to obtain more accurate reflectance parameters. The strip was illuminated at 45˚c through optical fiber and the intensity of reflected light was determined at rectanguLat angle using a photodiode. The kubelka-munk coefficient and reflection optical density were determined ar four different wavelengths(500, 550, 570 and 610nm) for blood glucose strip. For higher concentration than 300mg/41 about glucose, a saturation state of abforbance was observed at 500, 550 and 570nm. The correlation between glucose concentration and parameters was the best at 610nm. 30535 T00401030535 ^x Radiation-induced fibrosarcoma tumors were grown on the flanks of C3H mice. The mice were divided into two groups. One group was injected with Photofrin II, intravenously (2.5mg/kg body weight). The other group received no Photofrin II. Mice from both groups were irradialed for approximately 15 minutes at 100, 300, or 500 mW/cm2 with the argon (488nm/514.5 nm), dye(628nm) and gold vapor (pulsed 628 nm) laser light. A photosensitizer behaved as an added absorber. Under our experimental conditions, the presence of Photolfrin II increased surface temperature by at least 40% and the temperature rise due to 300 mW/cm2 irradiation exceeded values for hyperthermia. Light and temperature distributions with depth were estimated by a computer model. The model demonstrated the influence of wavelength on the thermal process and proved to be a valuable tool to investigate internal temperature rise. 30536 T00401030536 ^x We investigated the structural geometry of thirty-eight Korean femurs. The purpose of this study is to identify major geometrical differences between Korean femurs 3nd others that we believe belong to Caucasians so that we would be able to get insights into the femoral component design that fits Asians including Koreans. We utilized computerized tomography (CT) images of femurs extracted from cadavers. The CT images were transformed into bitmap data by using a film scanner, and then analyzed by using a commercially available software called Image v.1.0 and a Macintosh IIci computer.The resulting data were compared with already published data. The major results show that the geometry of the Korean femurs is significantly different from that of Caucasians: (1) the anteversion angle and the canal flare index are greater by the amount of approximately 8˚ and 0.5, respectively, (2) the shape of the isthmus cross section is more round, and (3) the distance between the teaser trochanter and the proximal border of the isthmus is shelter by about 15 mm. The results suggested that the femoral component suitable for Asians should be different from the currently-used components designed and manufactured mostly by European or American companies. 30537 T00401030537 ^x It is well known that nonlinear propagation characteristics of the wave in the tissue may give very useful information for the medical diagnoisis. In this paper, a new method to detect nonlinear propagation characteristics of the internal vibration in the tissue for the low frequency mechanical vibration by using bispectral analysis is proposed. In the method, low frequency vibration of f0( = 100Hz) is applied on the surface of the object, and the waveform of the internal vibration x (t) is measured from Doppler frequency modulation of silmultaneously transmitted probing ultrasonic waves. Then, the bispectra of the signal x (t) at the frequencies (f0, f0) and (f0, 2f0) are calculated to estimate the nonlinear propagation characteristics as their magnitude ratio, w here since bispectrum is free from the gaussian additive noise we can get the value with high S/N. Basic experimental system is constructed by using 3.0 MHz probing ultrasonic waves and the several experiments are carried out for some phantoms. Results show the superiority of the proposed method to the conventional method using power spectrum and also its usefulness for the tissue characterization. 30541 T00401030541 ^x This paper describes the implementation of a computerized radial pulse diagnosis by aids of a clinical expert. On this base, we composed of the radial pulse diagnosis system in korean traditional medicine. The system composed of a radial pulse wave detection system and a radial pulse diagnosis system. With a detection system, we detected Inyoung and Cheongu radial pulse wave and processed it. Then, we have got the characteristic parameters of radial pulse wave and also quantified that according to the method of Inyoung-Cheongu Comparison Radial Pulse Diagnosis. We defined the jugement standard of radial pulse diagnosis system and then we confirmed the possibility for realization of automatic radial pulse diagnosis in korean traditional medicine. 30545 T00401030545 ^x Microspheres are expected to be applied to biomedical areas such as solid-phase immunoassays, drug delivery systems, immunomagnetic cell separation. To synthesize microspheres for biomedical application, "two stage shot growth method" was developed. The uniformity ratio of synthesized microspheres was always smaller than 1.05. And the surface charge density (or the number of ionizable functional groups) of the microspheres synthesized by "two stage shot growth method" was 6~13 times higher than that of the microspheres synthesized by conventional seeded batch copolymerization. As a previous step for biomedical application, adsorption experiments of bovine albumin on microspheres were carried out under various conditions. The maximum adsorbed amount was obtained in the neighborhood of pH 4.5. Isoelectric point of bovine albumin is pH 5.0, so experimental result shows that it shifted to acid area. The adsorption isotherm was obtained, the plateau region was always reached at 2.Og/L (bulk concentration of bovine albumin).The effect of the kind and the amount of surface functional group was also examined. 30575 T00401030575 ^x A medical image workstation was developed using multimedia technique. The system based on PC-486DX was designed to acquire medical images produced by medical imaging instruments and related audio information, that is, doctors' reporting results. Input information was processed and analyzed, then the results were presented in the form of graph and animation. All the informations of the system were hierarchically related with the image as the apex. Processing and analysis algorithms were implemented so that the diagnostic accuracy could be improved. The diagnosed information can be transferred for patient diagnosis through LAN(local area network). 30592 T00401030592 ^x In the conventional infrared imaging system, complex infrared lens systems are usually used for directing collimated narrow infrared beams into the high speed 2-dimensional optic scanner. In this paper, a simple reflective infrared optic system with a 2-dimensional optic scanner is proposed for the realization of medical infrared thermography system. It has been experimentally proven that the intfrared thermography system composed of the proposed optic system has the temperature resolution of 0.1˚c under the spatial resolution of lmrad, the image matrix size of 256 X 240, and tile imaging time of 4 seconds. 30593 T00401030593 ^x In this paper, MIIS (Medical Image Information System) has been designed and implemented using INGRES RDBMS, which is based on a client/server architecture. The implemented system allows users to register and retrieve patient information, medical images and diagnostic reports. It also provides the function to display these information on workstation windows simultaneously by using the designed menu-driven graphic user interface. The medical image compression/decompression techniques are implemented and integrated into the medical image database system for the efficient data storage and the fast access through the network. 30594 T00401030594 ^x In this paper, computerized BEAM was implemented for the space domain analysis of EEG. Trans-formation from temporal summation to two-dimensional mappings is formed by 4 nearest point inter-polaton method. Methods of representation of BEAM are two. One is dot density method which classify brain electrical potential 9 levels by dot density of gray levels and the other is colour method which classify brain electrical 12 levels by red-green colours. In this BEAM, instantaneous change and average energy distribution over any arbitrary time interval of brain electrical activity could be observed and analyzed easily. In the frequency domain, the distribution of energy spectrum of a special band can easily be distinguished normality and abnormality. 30608 T00401030608 ^x Laboratory information system (LIS) is a key tool to manage laboratory data in clinical pathology. Our department has developed an information system for routine hematology using down-sized computer system. We have used an IBM 486 compatible PC with 16MB main memory, 210 MB hard disk drive, 9 RS-232C port and 24 pin dot printer. The operating system and database management system were SCO UNIX and SCO foxbase, respectively. For program development, we used Xbase language provided by SCO foxbase. The C language was used for interface purpose. To make the system use friendly, pull-down menu was used. The system connected to our hospital information system via application program interface (API), so the information related to patient and request details is automatically transmitted to our computer. Our system interfaced with fwd complete blood count analyzers(Sysmex NE-8000 and Coulter STKS) for unidirectional data tansmission from analyzer to computer. The authors suggests that this system based on down-sized computer could provide a progressive approach to total LIS based on local area network, and the implemented system could serve as a model for other hospital's LIS for routine hematology. 30609 T00401030609 ^x To develop an artificial bone substitute that is gradually degraded and replaced by the regenerated natural bone, the authors designed a composite that is consisted of calcium phosphate and collagen. To use as the structural matrix of the composite, collagen was purified from human umbilical cord. The obtained collagen was treated by pepsin to remove telopeptides, and finally, the immune-free atelocollagen was produced: The cross linked atelocollagen was highly resistant to the collagenase induced collagenolysis. The cross linked collagen demonstrated an improved tensile strength. 30618 T00401030618 ^x This paper is a study on the design of adptive filter for QRS complex detection. We propose a simple adaptive algorithm to increase capability of noise cancelation in QRS complex detection with two stage adaptive filter. At the first stage, background noise is removed and at the next stage, only spectrum of QRS complex components is passed. Two adaptive filters can afford to keep track of the changes of both noise and QRS complex. Each adaptive filter consists of prediction error filter and FIR filter The impulse response of FIR filter uses coefficients of prediction error filter. The detection rates for 105 and 108 of MIT/BIH data base were 99.3% and 97.4% respectively. 30619 T00401030619 ^x To develop an artificial bone substitute that is gradually degraded and replaced by the regenerated natural bone, the authors designed and produced a composite that is consisted of calcium phosphate and collagen. Human umbilical cord origin pepsin treated type I atelocollagen was used as the structural matrix, by which sintered or non-sintered carbonate apatite was encapsulated to form an inorganic-organic composite. With cross linking atelocollagen by UV ray irradiation, the resistance to both compressive and tensile strength was increased. Collagen degradation by the collagenase induced collagenolysis was also decreased. 30620 T00401030620 ^x We have developed a monoleaflet polymer valve as an inexpensive and viable alternative, especially for short-term use in the ventricular assist device or total artificial heart. The frame and leaflet of the polymer valve were made from polyurethane, To evaluate the hemodynamic performance of the polymer valve a comparative study of flow dynamics past a polymer valve and a St. Jude Medical prosthetic valve under physiological pulsatile flow conditions in vitro was made. Comparisons between the valves were made on the transvalvular pressure drop, regurgitation volume and maximum valve opening area. The polymer valve showed smaller regurgitation volume and transvalvular pressure drop compared to the mechanical valve at higher heart rate. The results showed that the functional characteristics of the polymer valve compared favorably with those of the mechanical valve at higher heart rate. 30621 T00401030621 ^x Explosive evaporative removal process of biological tissue by absorption of a CW laser has been simulated by using gelatin and a multimode Nd:YAG laser. Because the point of maximun temperature of laser-irradiated gelatin exists below the surface due to surface cooling, evaporation at the boiling temperature is made explosively from below the surface. The important parameters of this process are the conduction loss to laser power absorption (defined as the conduction-to-laser power parameter, Nk), the convection heat transfer at the surface to conduction loss (defined as Bi), dimensionless extinction coefficient (defined as Br.), and dimensionless irradiation time (defined as Fo). Dependence of Fo on Nk and Bi has been observed by experiment, and the results have been compared with the numerical results obtained by solving a 2-dimensional conduction equation. Fo and explosion depth (from the surface to the point of maximun temperature) are increased when Nk and Bi are increased.To find out the minimum laser power for explosive evaporative removal process, steady state analysis has been also made. The limit of Nk to induce evaporative removal, which is proportional to the inverse of the laser power, has been obtained. 30622 T00401030622 ^x N1 and N2 gross neural action potentials were measured from the round window of the guinea pig cochlea at the onset of the acoustic stimuli. N1-N2 audiograms were made by means of regulating stimulant intensities in order to produce constant N1-N2 potentials as criteria for different input tone pip frequencies. The lowest threshold was measured with an input tone pip I5 dB SPL in intensity and 12 KHz in frequency when the animal was in normal physiological condition. The procedure of experimental measurements is explained in detail. This experimental approach is very useful for the investigation of the Cochlear function. Both noN1inear and active functions of the Cochlea can be monitored by N1-N2 audiograms. 30623 T00401030623 ^x In electrical impedance tomography(EIT), we use boundary current and voltage measurements toprovide the information about the cross-sectional distribution of electrical impedance or resistivity. One of the major problems in EIT has been the inaccessibility of internal voltage or current data in finding the internal impedance values. We propose a new image reconstruction method using internal current density data measured by NMR. We obtained a two-dimensional current density distribution within a phantom by processing the real and imaginary MR images from a 4.77 NMR machine. We implemented a resistivity mage reconstruction algorithm using the finite element method and sensitivity matrix. We presented computer simulation results of the mage reconstruction algorithm and furture direction of the research. 30624 T00401030624 ^x A new method of digital image analysis technique for discrimination of cancer cell was presented in this paper. The object image was the Thyroid eland cells image that was diagnosed as normal and abnormal (two types of abnormal: follicular neoplastic cell, and papillary neoplastic cell), respectively. By using the proposed region segmentation algorithm, the cells were segmented into nucleus. The 16 feature parameters were used to calculate the features of each nucleus. A9 a consequence of using dominant feature parameters method proposed in this paper, discrimination rate of 91.11% was obtained for Thyroid Gland cells. 30625 T00401030625 ^x An electrical stimulator was designed to induce locomotion for paraplegic patients caused by central nervous system injury. Optimal stimulus parameters, which can minimize muscle fatigue and can achieve effective muscle contraction were determined in slow and fast muscles in Sprague-Dawley rats. Stimulus patterns of our stimulator were designed to simulate electromyographic activity monitored during locomotion of normal subjects. Muscle types of the lower extremity were classified according to their mechanical property of contraction, which are slow muscle (msoleus m.) and fast muscle (medial gastrocneminus m., rectus femoris m., vastus lateralis m.). Optimal parameters of electrical stimulation for slow muscles were 20 Hz, 0.2 ms square pulse. For fast muscle, 40 Hz, 0.3 ms square pulse was optimal to produce repeated contraction. Higher stimulus intensity was required when synergistic muscles were stimulated simultaneously than when they were stimulated individually. Electrical stimulation for each muscle was designed to generate bipedal locomotion, so that individual muscles alternate contraction and relaxation to simulate stance and swing phases. Portable electrical stimulator with 16 channels built in microprocessor was constructed and applied to paraplegic patients due to lumbar cord injury. The electrical stimulator restored partially gait function in paraplegic patients. 30626 T00401030626 ^x Two-Dimensional modelling of the Cochlear biomechanics is presented in this paper. The Laplace partial differential equation which represents the fluid mechanics of the Cochlea has been transformed into two-dimensional electrical transmission line. The procedure of this transformation is explained in detail. The comparison between one and two dimensional models is also presented. This electrical modelling of the basilar membrane (BM) is clearly useful for the next approach to the further. Development of active elements which are essential in the producing of the sharp tuning of the BM. This paper shows that two-dimension model is qualitatively better than one-dimensional model both in amplitude and phase responses of the BM displacement. The present model is only for frequency response. However because the model is electrical, the two-dimensional transmission line model can be extended to time response without any difficult. 30627 T00401030627 ^x A method has been proposed for the fully automatic detection of left ventricular endocardial boundary in 2D short axis echocardiogram using geometric model. The procedure has the following three distinct stages. First, the initial center is estimated by the initial center estimation algorithm which is applied to decimated image. Second, the center estimation algorithm is applied to original image and then best-fit elliptic model estimation is processed. Third, best-fit boundary is detected by the cost function which is based on the best-fit elliptic model. The proposed method shows effective result without manual intervention by a human operator. 30628 T00401030628 ^x The intelligent trajectory control method that controls moving direction and average velocity for a prosthetic arm is proposed by pattern recognition and force estimations using EMG signals. Also, we propose the real time trajectory planning method which generates continuous accelleration paths using 3 stage linear filters to minimize the impact to human body induced by arm motions and to reduce the muscle fatigue. We use combination of MLP and fuzzy filter for pattern recognition to estimate the direction of a muscle and Hogan's method for the force estimation. EMG signals are acquired by using a amputation simulator and 2 dimensional joystick motion. The simulation results of proposed prosthetic arm control system using the EMf signals show that the arm is effectively followed the desired trajectory depended on estimated force and direction of muscle movements. 30638 T00401030638 ^x A new neural network architecture for the recognition of patterns from images is proposed, which is partially based on the results of physiological studies. The proposed network is composed of multi-layers and the nerve cells in each layer are connected by spatial filters which approximate receptive fields in optic nerve fields. In the proposed method, patterns recognition for complicated images is carried out using global features as well as local features such as lines and end-points. A new generating method of matched filers representing global features is proposed in this network. 30659 T00401030659 ^x An implementation scheme of the magnetic nerve stimulator using a switching mode power supply is proposed. By using a switching mode power supply rather than a conventional linear power supply for charging high voltage capacitors, the weight and size of the magnetic nerve stimulator can be considerably reduced. Maximum output voltage of the developed magnetic nerve stimulator using the switching mode power supply is 3, 000 volts and switching time is about 100 msec. Experimental results or human nerve stimulations using the developed stimulator are presented. 30768 T00401030768 ^x In this paper, we describe the design methodology and specifications of the developed module-based bedside monitors for patient monitoring. The bedside monitor consists of a main unit and module cases with various parameter modules. The main unit includes a 12.1" TFT color LCD, a main CPU board, and peripherals such as a module controller, Ethernet LAN card, video card, rotate/push button controller, etc. The main unit can connect at maximum three module cases each of which can accommodate up to 7 parameter modules. They include the modules for electrocardiograph, respiration, invasive blood pressure, noninvasive blood pressure, temperature, and SpO2 with Plethysmograph.SpO2 with Plethysmograph.

• Journal of Intelligence and Information Systems
• /
• v.22 no.1
• /
• pp.139-157
• /
• 2016

Bankruptcy involves considerable costs, so it can have significant effects on a country's economy. Thus, bankruptcy prediction is an important issue. Over the past several decades, many researchers have addressed topics associated with bankruptcy prediction. Early research on bankruptcy prediction employed conventional statistical methods such as univariate analysis, discriminant analysis, multiple regression, and logistic regression. Later on, many studies began utilizing artificial intelligence techniques such as inductive learning, neural networks, and case-based reasoning. Currently, ensemble models are being utilized to enhance the accuracy of bankruptcy prediction. Ensemble classification involves combining multiple classifiers to obtain more accurate predictions than those obtained using individual models. Ensemble learning techniques are known to be very useful for improving the generalization ability of the classifier. Base classifiers in the ensemble must be as accurate and diverse as possible in order to enhance the generalization ability of an ensemble model. Commonly used methods for constructing ensemble classifiers include bagging, boosting, and random subspace. The random subspace method selects a random feature subset for each classifier from the original feature space to diversify the base classifiers of an ensemble. Each ensemble member is trained by a randomly chosen feature subspace from the original feature set, and predictions from each ensemble member are combined by an aggregation method. The k-nearest neighbors (KNN) classifier is robust with respect to variations in the dataset but is very sensitive to changes in the feature space. For this reason, KNN is a good classifier for the random subspace method. The KNN random subspace ensemble model has been shown to be very effective for improving an individual KNN model. The k parameter of KNN base classifiers and selected feature subsets for base classifiers play an important role in determining the performance of the KNN ensemble model. However, few studies have focused on optimizing the k parameter and feature subsets of base classifiers in the ensemble. This study proposed a new ensemble method that improves upon the performance KNN ensemble model by optimizing both k parameters and feature subsets of base classifiers. A genetic algorithm was used to optimize the KNN ensemble model and improve the prediction accuracy of the ensemble model. The proposed model was applied to a bankruptcy prediction problem by using a real dataset from Korean companies. The research data included 1800 externally non-audited firms that filed for bankruptcy (900 cases) or non-bankruptcy (900 cases). Initially, the dataset consisted of 134 financial ratios. Prior to the experiments, 75 financial ratios were selected based on an independent sample t-test of each financial ratio as an input variable and bankruptcy or non-bankruptcy as an output variable. Of these, 24 financial ratios were selected by using a logistic regression backward feature selection method. The complete dataset was separated into two parts: training and validation. The training dataset was further divided into two portions: one for the training model and the other to avoid overfitting. The prediction accuracy against this dataset was used to determine the fitness value in order to avoid overfitting. The validation dataset was used to evaluate the effectiveness of the final model. A 10-fold cross-validation was implemented to compare the performances of the proposed model and other models. To evaluate the effectiveness of the proposed model, the classification accuracy of the proposed model was compared with that of other models. The Q-statistic values and average classification accuracies of base classifiers were investigated. The experimental results showed that the proposed model outperformed other models, such as the single model and random subspace ensemble model.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.11
• /
• pp.281-290
• /
• 2020

With the spread of smartphones, services that want to use personalized data are increasing. In particular, healthcare-related services deal with a variety of data, and data visualization techniques are used to effectively show this. As data visualization techniques are used, interactions in visualization are also naturally emphasized. In the PC environment, since the interaction for data visualization is performed with a mouse, various filtering for data is provided. On the other hand, in the case of interaction in a mobile environment, the screen size is small and it is difficult to recognize whether or not the interaction is possible, so that only limited visualization provided by the app can be provided through a button touch method. In order to overcome the limitation of interaction in such a mobile environment, we intend to enable data visualization interactions through conversations with chatbots so that users can check individual data through various visualizations. To do this, it is necessary to convert the user's query into a query and retrieve the result data through the converted query in the database that is storing data periodically. There are many studies currently being done to convert natural language into queries, but research on converting user queries into queries based on visualization has not been done yet. Therefore, in this paper, we will focus on query generation in a situation where a data visualization technique has been determined in advance. Supported interactions are filtering on task x-axis values and comparison between two groups. The test scenario utilized data on the number of steps, and filtering for the x-axis period was shown as a bar graph, and a comparison between the two groups was shown as a line graph. In order to develop a natural language processing model that can receive requested information through visualization, about 15,800 training data were collected through a survey of 1,000 people. As a result of algorithm development and performance evaluation, about 89% accuracy in classification model and 99% accuracy in query generation model was obtained.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.1-17
• /
• 2017

The deep learning framework is software designed to help develop deep learning models. Some of its important functions include "automatic differentiation" and "utilization of GPU". The list of popular deep learning framework includes Caffe (BVLC) and Theano (University of Montreal). And recently, Microsoft's deep learning framework, Microsoft Cognitive Toolkit, was released as open-source license, following Google's Tensorflow a year earlier. The early deep learning frameworks have been developed mainly for research at universities. Beginning with the inception of Tensorflow, however, it seems that companies such as Microsoft and Facebook have started to join the competition of framework development. Given the trend, Google and other companies are expected to continue investing in the deep learning framework to bring forward the initiative in the artificial intelligence business. From this point of view, we think it is a good time to compare some of deep learning frameworks. So we compare three deep learning frameworks which can be used as a Python library. Those are Google's Tensorflow, Microsoft's CNTK, and Theano which is sort of a predecessor of the preceding two. The most common and important function of deep learning frameworks is the ability to perform automatic differentiation. Basically all the mathematical expressions of deep learning models can be represented as computational graphs, which consist of nodes and edges. Partial derivatives on each edge of a computational graph can then be obtained. With the partial derivatives, we can let software compute differentiation of any node with respect to any variable by utilizing chain rule of Calculus. First of all, the convenience of coding is in the order of CNTK, Tensorflow, and Theano. The criterion is simply based on the lengths of the codes and the learning curve and the ease of coding are not the main concern. According to the criteria, Theano was the most difficult to implement with, and CNTK and Tensorflow were somewhat easier. With Tensorflow, we need to define weight variables and biases explicitly. The reason that CNTK and Tensorflow are easier to implement with is that those frameworks provide us with more abstraction than Theano. We, however, need to mention that low-level coding is not always bad. It gives us flexibility of coding. With the low-level coding such as in Theano, we can implement and test any new deep learning models or any new search methods that we can think of. The assessment of the execution speed of each framework is that there is not meaningful difference. According to the experiment, execution speeds of Theano and Tensorflow are very similar, although the experiment was limited to a CNN model. In the case of CNTK, the experimental environment was not maintained as the same. The code written in CNTK has to be run in PC environment without GPU where codes execute as much as 50 times slower than with GPU. But we concluded that the difference of execution speed was within the range of variation caused by the different hardware setup. In this study, we compared three types of deep learning framework: Theano, Tensorflow, and CNTK. According to Wikipedia, there are 12 available deep learning frameworks. And 15 different attributes differentiate each framework. Some of the important attributes would include interface language (Python, C ++, Java, etc.) and the availability of libraries on various deep learning models such as CNN, RNN, DBN, and etc. And if a user implements a large scale deep learning model, it will also be important to support multiple GPU or multiple servers. Also, if you are learning the deep learning model, it would also be important if there are enough examples and references.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.34 no.1
• /
• pp.66-74
• /
• 2005

This survey was conducted to assess the consumer perceptions of food-related hazard in 500 housewives from all over Korea. The subjects were selected by stratified random sampling method. The survey was performed using structured questionnaire through telephone interview by skilled interviewers. The results showed that 34.6% of the respondents felt secure and were not concerned about food safety, and 65.4% were concerned about food safety. Logistic regression analysis showed that the increasing concern on food brands, food additives (such as food preservatives and artificial color), and imported foodstuffs indicated the current increasing concern on food safety. Other related factors indicating the increasing concern on food safety were education level and care for children's health. The respondents who cared about food safety expressed a high degree of concern on processed foodstuffs such as commercial boxed lunch (93.3%), imported foods (92.7%), fastfoods (89.9%), processed meat products (88.7%), dining out (85.6%), cannery and frozen foods (83.5%), and instant foods (82.0%). The lowest degree of concern was on rice. All the respondents perceived that residues of chemical substances such as pesticides and food additives, and endocrine disrupters were the most potential food risk factors, followed by food-borne pathogens, and GMOs (Genetically Modified Organisms). However, these results were not consistent with scientific judgment. Therefore, more education and information were needed for consumers' awareness of facts and myths about food safety. In addition, the results showed that consumers put lower trust in food products information such as food labels, cultivation methods (organic or not), quality labels, and the place of origin. Nevertheless, the respondents expressed their desire to overcome alienation, and recognized the importance of knowing of the origin or the producers of food. They identified that people who need to take extreme precautions on food contamination were the producers, government officials, food companies, consumers, the consumer's association, and marketers, arranged in the order of highest to lowest. They also believed that the production stage of agriculture was the most important step for improving the level of food safety Therefore, the results indicated that there is a need to introduce safety systems in the production of agricultural products, as follows: Good Agricultural Practice (GAP), Hazard Analysis and Critical Control Point (HACCP), and Traceability System (75).

• Journal of Preventive Medicine and Public Health
• /
• v.27 no.4 s.48
• /
• pp.711-734
• /
• 1994

Among chemical agents in herbicides, dioxin (2, 3, 7, 8-tetrachloro dibenzo-$\rho$-dioxin : TCDD), a chemical contaminant in herbicides sprayed during the Vietnam War has been known to be the major agent causing toxic effects. Approximately 320,000 korean soldiers participated the Vietnam War from 1964 to 1974. Although the potential hazards of the herbicides among Korean veterans exposed were implicated, the problem had not been a public issue until 1991 when Korean veterans were informed U.S. companies, the herbicides manufacturer payed fund, from which a trust fund for New Zealand and Australian Class members were established in 1985. After a series of appeals and demonstration by the Korean Veterans demanding medical care and compensation for their serious health damages, a bill of medical care and compensation for herbicides victims was promulgated in March 1993 and become effective from May 1993, This study was carried out with two major objectives : the first to understand the health problems caused from the herbicides by reviewing literatures published, and the second to examine the nature and extent of health impacts among Korean veterans exposed and to develop valid study methods for the major study by interviewing and reviewing records on a part of veterans (638 persons) registered and completed medical examination in Seoul Veterans Administration Hospital from June to October 1993. The results obtained are as followings: 1. The literature review of 107 papers revealed that 1) Dioxin is teratogenic, carcinogenic and affects almost all organs including nervous, endocrine, and reproductive systems in animal experiments. 2) The diseases showing evidence of causal association were Hodgkin's disease, non-Hodgkin's disease, lung cancer, lymphoma, soft tissue sarcoma, chloroacne and polyneuropathy when judged on the basis of consistency in study results and biological plausibility. 2. Interview and medical record review study on 638 veterans, though limited validity owing to lack of control group, crude estimates of dioxin exposure levels (no biomarkers measurable), and uncertainty of diagnosis, showed that: 1) Most of the study subject's were in their 40's of age and had been dispatched to Vietnam during the period from $1965{\sim}1970$ around one year. 2) Most frequently complained symptoms in medical examination were motor weakness (32%), sensory abnormalities in extremities (23%), skin diseases (22%), and pain in extremities (20%) whereas in Interview they were more frequent in order of skin problem (44%), motor weakness (38%), sensory abnormalities and pain in extremities(17% and 19% each). Kappa indices on the same category of complaints between two sources of information were variable and relatively low. 3) On medical examination, only a part of the 638 subjects had initial impression (442 pts) and final diagnosis (218 pts) suggesting decision making on diagnosis appeared to be difficult even with all available modern medical technologies: in initial impression disorders from peripheral and central neuropathy were predominant whereas in final diagnosis various types of skin disorder were most frequent 4) When dose-response relationship between several conditions (from questionnaire) and arbitrary exposure scores were examined by CMH linear trend test, spontaneous abortion, sexual problems and health problem of offsprings showed statistically significant linear trends. However, pregnancy, accident and suicidal attempts did not show any relationship in this study capacity. 5) Among complaints, psychosis and neurosis (anxiety, phobia) in interview study, and memory disorder and psychosis in medical record study revealed linear trend. 6) Skin disorder was the only condition showing linear trend in initial impression and none in final diagnosis on medical examination. Even though objective to select out dioxin-related disease or group of diseases from this study was not achieved the research experiences provided firm basis for developing various methodological approaches. 3. From this preliminary study we concluded that a larger scale major epidemiologic study on health impacts of herbicides among Korean veterans exposed is not only Indispensible but also well designed study with more valid exposure information and diagnosis may be able to establish causal relationship between certain groups of diseases and exposure to the herbicides among Korean veterans.

• Korean journal of communication and information
• /
• v.35
• /
• pp.250-291
• /
• 2006

This study examines the regulatory issues and introduction problems of TV-betting data broadcasts in Korea by in-depth interview with a panel group. TV-betting data broadcast services of card games and horse racing games are widely in use in Europe and other parts of the world. In order to carry out the study, a demo program of TV-betting data broadcast in the OCAP(OpenCableTM Application Platform Specification) system environment, which is the data broadcasting standard for digital cable broadcasts in Korea was exposed to the panel group and then they were interviewed after watching and using the program. The results could be summarized as below. First of all, while TV-betting data broadcasts have many elements of entertainment, the respondents thought that it would be difficult to introduce TV-betting in data broadcasts as in overseas countries largely due to social factors. In addition, in order to introduce TV-betting data broadcasts, they suggested that excessive speculativeness must be suppressed through a series of regulatory system devices, such as by guaranteeing credibility of the media based on safe security systems for transactions, scheduling programs with effective time constraints to prevent the games from running too frequently, limiting the betting values, and by prohibiting access to games through set-top boxes of other data broadcast subscribers. The general consensus was that TV-betting could be considered for gradual introduction within the governmental laws and regulations that would minimize its ill effects. Therefore, the government should formulate long-term regulations and policies for data broadcasts. Once the groundwork is laid for safe introduction of TV-betting on data broadcasts within the boundary of laws and regulations, interactive TV games are expected to be introduced in Korea not only for added functionality of entertainment but also for far-ranging development of data broadcast and new media industries.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.6
• /
• pp.1210-1230
• /
• 1997

In this paper, we suggest an architecture of DS/CDMA tranceiver composed of one pilot channel used as reference and multiple traffic channels. The pilot channel-an unmodulated PN code-is used as the reference signal for synchronization of PN code and data demondulation. The coherent demodulation architecture is also exploited for the reverse link as well as for the forward link. Here are the characteristics of the suggested DS/CDMA system. First, we suggest an interlaced quadrature spreading(IQS) method. In this method, the PN coe for I-phase 1st channel is used for Q-phase 2nd channels and the PN code for Q-phase 1st channel is used for I-phase 2nd channel, and so on-which is quite different from the eisting spreading schemes of DS/CDMA systems, such as IS-95 digital CDMA cellular or W-CDMA for PCS. By doing IQS spreading, we can drastically reduce the zero crossing rate of the RF signals. Second, we introduce an adaptive threshold setting for the synchronization of PN code, an initial acquistion method that uses a single PN code generator and reduces the acquistion time by a half compared the existing ones, and exploit the state machines to reduce the reacquistion time Third, various kinds of functions, such as automatic frequency control(AFC), automatic level control(ALC), bit-error-rate(BER) estimator, and spectral shaping for reducing the adjacent channel interference, are introduced to improve the system performance. Fourth, we designed and implemented the DS/CDMA MODEM to be used for variable transmission rate applications-from 16Kbps to 1.024Mbps. We developed and confirmed the DS/CDMA MODEM architecture through mathematical analysis and various kind of simulations. The ASIC design was done using VHDL coding and synthesis. To cope with several different kinds of applications, we developed transmitter and receiver ASICs separately. While a single transmitter or receiver ASC contains three channels (one for the pilot and the others for the traffic channels), by combining several transmitter ASICs, we can expand the number of channels up to 64. The ASICs are now under use for implementing a line-of-sight (LOS) radio equipment.