• Title/Summary/Keyword: Mass Memory

Search Result 183, Processing Time 0.026 seconds

A Design on Informal Big Data Topic Extraction System Based on Spark Framework (Spark 프레임워크 기반 비정형 빅데이터 토픽 추출 시스템 설계)

  • Park, Kiejin
    • KIPS Transactions on Software and Data Engineering
    • /
    • v.5 no.11
    • /
    • pp.521-526
    • /
    • 2016
  • As on-line informal text data have massive in its volume and have unstructured characteristics in nature, there are limitations in applying traditional relational data model technologies for data storage and data analysis jobs. Moreover, using dynamically generating massive social data, social user's real-time reaction analysis tasks is hard to accomplish. In the paper, to capture easily the semantics of massive and informal on-line documents with unsupervised learning mechanism, we design and implement automatic topic extraction systems according to the mass of the words that consists a document. The input data set to the proposed system are generated first, using N-gram algorithm to build multiple words to capture the meaning of the sentences precisely, and Hadoop and Spark (In-memory distributed computing framework) are adopted to run topic model. In the experiment phases, TB level input data are processed for data preprocessing and proposed topic extraction steps are applied. We conclude that the proposed system shows good performance in extracting meaningful topics in time as the intermediate results come from main memories directly instead of an HDD reading.

Deep Learning-Based Vehicle Anomaly Detection by Combining Vehicle Sensor Data (차량 센서 데이터 조합을 통한 딥러닝 기반 차량 이상탐지)

  • Kim, Songhee;Kim, Sunhye;Yoon, Byungun
    • Journal of the Korea Academia-Industrial cooperation Society
    • /
    • v.22 no.3
    • /
    • pp.20-29
    • /
    • 2021
  • In the Industry 4.0 era, artificial intelligence has attracted considerable interest for learning mass data to improve the accuracy of forecasting and classification. On the other hand, the current method of detecting anomalies relies on traditional statistical methods for a limited amount of data, making it difficult to detect accurate anomalies. Therefore, this paper proposes an artificial intelligence-based anomaly detection methodology to improve the prediction accuracy and identify new data patterns. In particular, data were collected and analyzed from the point of view that sensor data collected at vehicle idle could be used to detect abnormalities. To this end, a sensor was designed to determine the appropriate time length of the data entered into the forecast model, compare the results of idling data with the overall driving data utilization, and make optimal predictions through a combination of various sensor data. In addition, the predictive accuracy of artificial intelligence techniques was presented by comparing Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM) as the predictive methodologies. According to the analysis, using idle data, using 1.5 times of the data for the idling periods, and using CNN over LSTM showed better prediction results.

Web Monitoring based Encryption Web Traffic Attack Detection System (웹 모니터링 기반 암호화 웹트래픽 공격 탐지 시스템)

  • Lee, Seokwoo;Park, Soonmo;Jung, Hoekyung
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.25 no.3
    • /
    • pp.449-455
    • /
    • 2021
  • This paper proposes an encryption web transaction attack detection system based on the existing web application monitoring system. Although there was difficulty in detecting attacks on the encrypted web traffic because the existing web traffic security systems detect and defend attacks based on encrypted packets in the network area of the encryption section between the client and server, by utilizing the technology of the web application monitoring system, it is possible to detect various intelligent cyber-attacks based on information that is already decrypted in the memory of the web application server. In addition, since user identification is possible through the application session ID, statistical detection of attacks such as IP tampering attacks, mass web transaction call users, and DDoS attacks are also possible. Thus, it can be considered that it is possible to respond to various intelligent cyber attacks hidden in the encrypted traffic by collecting and detecting information in the non-encrypted section of the encrypted web traffic.

Force-deformation relationship prediction of bridge piers through stacked LSTM network using fast and slow cyclic tests

  • Omid Yazdanpanah;Minwoo Chang;Minseok Park;Yunbyeong Chae
    • Structural Engineering and Mechanics
    • /
    • v.85 no.4
    • /
    • pp.469-484
    • /
    • 2023
  • A deep recursive bidirectional Cuda Deep Neural Network Long Short Term Memory (Bi-CuDNNLSTM) layer is recruited in this paper to predict the entire force time histories, and the corresponding hysteresis and backbone curves of reinforced concrete (RC) bridge piers using experimental fast and slow cyclic tests. The proposed stacked Bi-CuDNNLSTM layers involve multiple uncertain input variables, including horizontal actuator displacements, vertical actuators axial loads, the effective height of the bridge pier, the moment of inertia, and mass. The functional application programming interface in the Keras Python library is utilized to develop a deep learning model considering all the above various input attributes. To have a robust and reliable prediction, the dataset for both the fast and slow cyclic tests is split into three mutually exclusive subsets of training, validation, and testing (unseen). The whole datasets include 17 RC bridge piers tested experimentally ten for fast and seven for slow cyclic tests. The results bring to light that the mean absolute error, as a loss function, is monotonically decreased to zero for both the training and validation datasets after 5000 epochs, and a high level of correlation is observed between the predicted and the experimentally measured values of the force time histories for all the datasets, more than 90%. It can be concluded that the maximum mean of the normalized error, obtained through Box-Whisker plot and Gaussian distribution of normalized error, associated with unseen data is about 10% and 3% for the fast and slow cyclic tests, respectively. In recapitulation, it brings to an end that the stacked Bi-CuDNNLSTM layer implemented in this study has a myriad of benefits in reducing the time and experimental costs for conducting new fast and slow cyclic tests in the future and results in a fast and accurate insight into hysteretic behavior of bridge piers.

Brand Equity and Purchase Intention in Fashion Products: A Cross-Cultural Study in Asia and Europe (상표자산과 구매의도와의 관계에 관한 국제비교연구 - 아시아와 유럽의 의류시장을 중심으로 -)

  • Kim, Kyung-Hoon;Ko, Eun-Ju;Graham, Hooley;Lee, Nick;Lee, Dong-Hae;Jung, Hong-Seob;Jeon, Byung-Joo;Moon, Hak-Il
    • Journal of Global Scholars of Marketing Science
    • /
    • v.18 no.4
    • /
    • pp.245-276
    • /
    • 2008
  • Brand equity is one of the most important concepts in business practice as well as in academic research. Successful brands can allow marketers to gain competitive advantage (Lassar et al.,1995), including the opportunity for successful extensions, resilience against competitors' promotional pressures, and the ability to create barriers to competitive entry (Farquhar, 1989). Branding plays a special role in service firms because strong brands increase trust in intangible products (Berry, 2000), enabling customers to better visualize and understand them. They reduce customers' perceived monetary, social, and safety risks in buying services, which are obstacles to evaluating a service correctly before purchase. Also, a high level of brand equity increases consumer satisfaction, repurchasing intent, and degree of loyalty. Brand equity can be considered as a mixture that includes both financial assets and relationships. Actually, brand equity can be viewed as the value added to the product (Keller, 1993), or the perceived value of the product in consumers' minds. Mahajan et al. (1990) claim that customer-based brand equity can be measured by the level of consumers' perceptions. Several researchers discuss brand equity based on two dimensions: consumer perception and consumer behavior. Aaker (1991) suggests measuring brand equity through price premium, loyalty, perceived quality, and brand associations. Viewing brand equity as the consumer's behavior toward a brand, Keller (1993) proposes similar dimensions: brand awareness and brand knowledge. Thus, past studies tend to identify brand equity as a multidimensional construct consisted of brand loyalty, brand awareness, brand knowledge, customer satisfaction, perceived equity, brand associations, and other proprietary assets (Aaker, 1991, 1996; Blackston, 1995; Cobb-Walgren et al., 1995; Na, 1995). Other studies tend to regard brand equity and other brand assets, such as brand knowledge, brand awareness, brand image, brand loyalty, perceived quality, and so on, as independent but related constructs (Keller, 1993; Kirmani and Zeithaml, 1993). Walters(1978) defined information search as, "A psychological or physical action a consumer takes in order to acquire information about a product or store." But, each consumer has different methods for informationsearch. There are two methods of information search, internal and external search. Internal search is, "Search of information already saved in the memory of the individual consumer"(Engel, Blackwell, 1982) which is, "memory of a previous purchase experience or information from a previous search."(Beales, Mazis, Salop, and Staelin, 1981). External search is "A completely voluntary decision made in order to obtain new information"(Engel & Blackwell, 1982) which is, "Actions of a consumer to acquire necessary information by such methods as intentionally exposing oneself to advertisements, taking to friends or family or visiting a store."(Beales, Mazis, Salop, and Staelin, 1981). There are many sources for consumers' information search including advertisement sources such as the internet, radio, television, newspapers and magazines, information supplied by businesses such as sales people, packaging and in-store information, consumer sources such as family, friends and colleagues, and mass media sources such as consumer protection agencies, government agencies and mass media sources. Understanding consumers' purchasing behavior is a key factor of a firm to attract and retain customers and improving the firm's prospects for survival and growth, and enhancing shareholder's value. Therefore, marketers should understand consumer as individual and market segment. One theory of consumer behavior supports the belief that individuals are rational. Individuals think and move through stages when making a purchase decision. This means that rational thinkers have led to the identification of a consumer buying decision process. This decision process with its different levels of involvement and influencing factors has been widely accepted and is fundamental to the understanding purchase intention represent to what consumers think they will buy. Brand equity is not only companies but also very important asset more than product itself. This paper studies brand equity model and influencing factors including information process such as information searching and information resources in the fashion market in Asia and Europe. Information searching and information resources are influencing brand knowledge that influences consumers purchase decision. Nine research hypotheses are drawn to test the relationships among antecedents of brand equity and purchase intention and relationships among brand knowledge, brand value, brand attitude, and brand loyalty. H1. Information searching influences brand knowledge positively. H2. Information sources influence brand knowledge positively. H3. Brand knowledge influences brand attitude. H4. Brand knowledge influences brand value. H5. Brand attitude influences brand loyalty. H6. Brand attitude influences brand value. H7. Brand loyalty influences purchase intention. H8. Brand value influence purchase intention. H9. There will be the same research model in Asia and Europe. We performed structural equation model analysis in order to test hypotheses suggested in this study. The model fitting index of the research model in Asia was $X^2$=195.19(p=0.0), NFI=0.90, NNFI=0.87, CFI=0.90, GFI=0.90, RMR=0.083, AGFI=0.85, which means the model fitting of the model is good enough. In Europe, it was $X^2$=133.25(p=0.0), NFI=0.81, NNFI=0.85, CFI=0.89, GFI=0.90, RMR=0.073, AGFI=0.85, which means the model fitting of the model is good enough. From the test results, hypotheses were accepted. All of these hypotheses except one are supported. In Europe, information search is not an antecedent of brand knowledge. This means that sales of global fashion brands like jeans in Europe are not expanding as rapidly as in Asian markets such as China, Japan, and South Korea. Young consumers in European countries are not more brand and fashion conscious than their counter partners in Asia. The results have theoretical, practical meaning and contributions. In the fashion jeans industry, relatively few studies examining the viability of cross-national brand equity has been studied. This study provides insight on building global brand equity and suggests information process elements like information search and information resources are working differently in Asia and Europe for fashion jean market.

  • PDF

A Case of Paraneoplastic Limbic Encephalitis Associated with Small Cell Lung Cancer

  • Ryu, Ja Young;Lee, Seung Hyeun;Lee, Eun Joo;Min, Kyung Hoon;Hur, Gyu Young;Lee, Sung Yong;Kim, Je Hyeong;Lee, Sang Yeub;Shin, Chol;Shim, Jae Jeong;In, Kwang Ho;Kang, Kyung Ho;Yoo, Se Hwa
    • Tuberculosis and Respiratory Diseases
    • /
    • v.73 no.5
    • /
    • pp.273-277
    • /
    • 2012
  • Paraneoplastic limbic encephalitis (PLE) is a rare syndrome characterized by memory impairment, affective and behavioral disturbances and seizures. Among many different neoplasms known to cause PLE, small cell lung cancer (SCLC) is the most frequently reported. The pathogenesis is not fully understood but is believed to be autoimmune-related. We experienced a patient with typical clinical features of PLE. A 67-year-old man presented with seizure and disorientation. Brain magnetic resonance imaging demonstrated high signal intensity in the bilateral amygdala and hippocampus in flair and T2-weighted images suggestive of limbic encephalitis. Cerebrospinal fluid tapping revealed no evidence of malignant cells or infection. Positron emission tomography/computed tomography showed a lung mass with pleural effusion and a consequent biopsy confirmed the diagnosis of PLE associated with SCLC. The patient was subsequently treated with chemotherapy and neurologic symptoms gradually improved.

A Study on the etching mechanism of $CeO_2$ thin film by high density plasma (고밀도 플라즈마에 의한 $CeO_2$ 박막의 식각 메커니즘 연구)

  • Oh, Chang-Seok;Kim, Chang-Il
    • Journal of the Institute of Electronics Engineers of Korea SD
    • /
    • v.38 no.12
    • /
    • pp.8-13
    • /
    • 2001
  • Cerium oxide ($CeO_2$) thin film has been proposed as a buffer layer between the ferroelectric thin film and the Si substrate in Metal-Ferroelectric-Insulator-Silicon (MFIS) structures for ferroelectric random access memory (FRAM) applications. In this study, $CeO_2$ thin films were etched with $Cl_2$/Ar gas mixture in an inductively coupled plasma (ICP). Etch properties were measured for different gas mixing ratio of $Cl_2$($Cl_2$+Ar) while the other process conditions were fixed at RF power (600 W), dc bias voltage (-200 V), and chamber pressure (15 mTorr). The highest etch rate of $CeO_2$ thin film was 230 ${\AA}$/min and the selectivity of $CeO_2$ to $YMnO_3$ was 1.83 at $Cl_2$($Cl_2$+Ar gas mixing ratio of 0.2. The surface reaction of the etched $CeO_2$ thin films was investigated using x-ray photoelectron spectroscopy (XPS) analysis. There is a Ce-Cl bonding by chemical reaction between Ce and Cl. The results of secondary ion mass spectrometer (SIMS) analysis were compared with the results of XPS analysis and the Ce-Cl bonding was monitored at 176.15 (a.m.u). These results confirm that Ce atoms of $CeO_2$ thin films react with chlorine and a compound such as CeCl remains on the surface of etched $CeO_2$ thin films. These products can be removed by Ar ion bombardment.

  • PDF

A study on Etch Characteristics of {Y-2}{O_3}$ Thin Films in Inductively Coupled Plasma (유도 결합 플라즈마를 이용한 {Y-2}{O_3}$ 박막의 식각 특성 연구)

  • Kim, Yeong-Chan;Kim, Chang-Il
    • Journal of the Institute of Electronics Engineers of Korea SD
    • /
    • v.38 no.9
    • /
    • pp.611-615
    • /
    • 2001
  • Y$_2$O$_3$ thin films have been proposed as a buffering insulator of metal/ferroelectric/insulator/semiconductor field effect transistor(MFISFET)-type ferroelectric random access memory (FRAM). In this study, $Y_2$O$_3$ thin films were etched with inductively coupled plasma(ICP). The etch rates of $Y_2$O$_3$ and YMnO$_3$, and the selectivity of $Y_2$O$_3$ to YMnO$_3$ were investigated by varying Cl$_2$/(Cl$_2$+Ar) gas mixing ratio. The maximum etch rate of $Y_2$O$_3$, and the selectivity of $Y_2$O$_3$ to YMnO$_3$ were 302$\AA$/min, and 2.4 at Cl$_2$/(Cl$_2$+Ar) gas mixing ratio of 0.2 respectively. Optical emission spectroscopy(OES) was used to understand the effects of gas combination on the etch rate of $Y_2$O$_3$ thin film. The surface reaction of the etched $Y_2$O$_3$ thin films was investigated by x-ray photoelectron spectroscopy (XPS). XPS analysis confirmed that there was chemical reaction between Y and Cl. This result was confirmed by secondary ion mass spectroscopy(SIMS) analysis.

  • PDF

A Multi-Wavelength Study of Galaxy Transition in Different Environments (다파장 관측 자료를 이용한 다양한 환경에서의 은하 진화 연구)

  • Lee, Gwang-Ho
    • The Bulletin of The Korean Astronomical Society
    • /
    • v.43 no.1
    • /
    • pp.34.2-35
    • /
    • 2018
  • Galaxy transition from star-forming to quiescent, accompanied with morphology transformation, is one of the key unresolved issues in extragalactic astronomy. Although several environmental mechanisms have been proposed, a deeper understanding of the impact of environment on galaxy transition still requires much exploration. My Ph.D. thesis focuses on which environmental mechanisms are primarily responsible for galaxy transition in different environments and looks at what happens during the transition phase using multi-wavelength photometric/spectroscopic data, from UV to mid-infrared (MIR), derived from several large surveys (GALEX, SDSS, and WISE) and our GMOS-North IFU observations. Our multi-wavelength approach provides new insights into the *late* stages of galaxy transition with a definition of the MIR green valley different from the optical green valley. I will present highlights from three areas in my thesis. First, through an in-depth study of environmental dependence of various properties of galaxies in a nearby supercluster A2199 (Lee et al. 2015), we found that the star formation of galaxies is quenched before the galaxies enter the MIR green valley, which is driven mainly by strangulation. Then, the morphological transformation from late- to early-type galaxies occurs in the MIR green valley. The main environmental mechanisms for the morphological transformation are galaxy-galaxy mergers and interactions that are likely to happen in high-density regions such as galaxy groups/clusters. After the transformation, early-type MIR green valley galaxies keep the memory of their last star formation for several Gyr until they move on to the next stage for completely quiescent galaxies. Second, compact groups (CGs) of galaxies are the most favorable environments for galaxy interactions. We studied MIR properties of galaxies in CGs and their environmental dependence (Lee et al. 2017), using a sample of 670 CGs identified using a friends-of-friends algorithms. We found that MIR [3.4]-[12] colors of CG galaxies are, on average, bluer than those of cluster galaxies. As CGs are located in denser regions, they tend to have larger early-type galaxy fractions and bluer MIR color galaxies. These trends can also be seen for neighboring galaxies around CGs. However, CG members always have larger early-type fractions and bluer MIR colors than their neighboring galaxies. These results suggest that galaxy evolution is faster in CGs than in other environments and that CGs are likely to be the best place for pre-processing. Third, post-starburst galaxies (PSBs) are an ideal laboratory to investigate the details of the transition phase. Their spectra reveal a phase of vigorous star formation activity, which is abruptly ended within the last 1 Gyr. Numerical simulations predict that the starburst, and thus the current A-type stellar population, should be localized within the galaxy's center (< kpc). Yet our GMOS IFU observations show otherwise; all five PSBs in our sample have Hdelta absorption line profiles that extend well beyond the central kpc. Most interestingly, we found a negative correlation between the Hdelta gradient slopes and the fractions of the stellar mass produced during the starburst, suggesting that stronger starbursts are more centrally-concentrated. I will discuss the results in relation with the origin of PSBs.

  • PDF

Work Hours and Cognitive Function: The Multi-Ethnic Study of Atherosclerosis

  • Charles, Luenda E.;Fekedulegn, Desta;Burchfiel, Cecil M.;Fujishiro, Kaori;Hazzouri, Adina Zeki Al;Fitzpatrick, Annette L.;Rapp, Stephen R.
    • Safety and Health at Work
    • /
    • v.11 no.2
    • /
    • pp.178-186
    • /
    • 2020
  • Background: Cognitive impairment is a public health burden. Our objective was to investigate associations between work hours and cognitive function. Methods: Multi-Ethnic Study of Atherosclerosis (MESA) participants (n = 2,497; 50.7% men; age range 44-84 years) reported hours per week worked in all jobs in Exams 1 (2000-2002), 2 (2002-2004), 3 (2004-2005), and 5 (2010-2011). Cognitive function was assessed (Exam 5) using the Cognitive Abilities Screening Instrument (version 2), a measure of global cognitive functioning; the Digit Symbol Coding, a measure of processing speed; and the Digit Span test, a measure of attention and working memory. We used a prospective approach and linear regression to assess associations for every 10 hours of work. Results: Among all participants, associations of hours worked with cognitive function of any type were not statistically significant. In occupation-stratified analyses (interaction p = 0.051), longer work hours were associated with poorer global cognitive function among Sales/Office and blue-collar workers, after adjustment for age, sex, physical activity, body mass index, race/ethnicity, educational level, annual income, history of heart attack, diabetes, apolipoprotein E-epsilon 4 allele (ApoE4) status, birth-place, number of years in the United States, language spoken at MESA Exam 1, and work hours at Exam 5 (β = -0.55, 95% CI = -0.99, -0.09) and (β = -0.80, -1.51, -0.09), respectively. In occupation-stratified analyses (interaction p = 0.040), we also observed an inverse association with processing speed among blue-collar workers (adjusted β = -0.80, -1.52, -0.07). Sex, race/ethnicity, and ApoE4 did not significantly modify associations between work hours and cognitive function. Conclusion: Weak inverse associations were observed between work hours and cognitive function among Sales/Office and blue-collar workers.