• Smart Media Journal
• /
• v.10 no.2
• /
• pp.76-83
• /
• 2021

With the development and use of smart devices such as smartphones and tablets increases, the mobile application market based on mobile devices is growing rapidly. Mobile application users write reviews to share their experience in using the application, which can identify consumers' various needs and application developers can receive useful feedback on improving the application through reviews written by consumers. However, there is a need to come up with measures to minimize the amount of time and expense that consumers have to pay to manually analyze the large amount of reviews they leave. In this work, we propose to collect delivery application user reviews from Google PlayStore and then use machine learning and deep learning techniques to classify them into four categories like application feature advantages, disadvantages, feature improvement requests and bug report. In the case of the performance of the Hugging Face's pretrained BERT-based Transformer model, the f1 score values for the above four categories were 0.93, 0.51, 0.76, and 0.83, respectively, showing superior performance than LSTM and GRU.

• Tunnel and Underground Space
• /
• v.11 no.4
• /
• pp.289-300
• /
• 2001

In the design of tunnel, it contains inaccuracy of data, fuzziness of evaluation, observer error and so on. The face observation during tunnel excavation, therefore, plays an important role to raise stability and to reduce supporting cost. This study is carried out to minimize the subjectiveness of observer and to exactly evaluate the natural properties of ground during the face observation. For these purpose, fuzzy set theory and neuro-fuzzy techniques in artificial intelligent techniques are applied to the inference of the RMR value from the observation data. The correlation between original RMR vague and inferred RM $R_{_FU}$ and RM $R_{_NF}$ values from fuzzy set theory and neuro-fuzzy techniques is investigated using 46 data. The results show that good correlation between original RMR value and infected RM $R_{_FU}$ and RM $R_{_NF}$ value is observed when the correlation coefficients are ｜R｜=0.96 and ｜R｜=0.95 respectively. From these results, applicability of fuzzy set theory and neuro-fuzzy techniques to rock mats classification is proved to be sufficiently high enough. enough.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.18
• /
• pp.8595-8598
• /
• 2016

Background: Decision making modalities for screening for many cancer conditions and different stages have become increasingly complex. Computer-based risk assessment systems facilitate scheduling and decision making and support the delivery of cancer screening services. The aim of this article was to survey electronic risk assessment system as an appropriate tool for the prevention of cancer. Materials and Methods: A qualitative design was used involving 21 face-to-face interviews. Interviewing involved asking questions and getting answers from exclusive managers of cancer screening. Of the participants 6 were female and 15 were male, and ages ranged from 32 to 78 years. The study was based on a grounded theory approach and the tool was a semi-structured interview. Results: Researchers studied 5 dimensions, comprising electronic guideline standards of colorectal cancer screening, work flow of clinical and genetic activities, pathways of colorectal cancer screening and functionality of computer based guidelines and barriers. Electronic guideline standards of colorectal cancer screening were described in the s3 categories of content standard, telecommunications and technical standards and nomenclature and classification standards. According to the participations' views, workflow and genetic pathways of colorectal cancer screening were identified. Conclusions: The study demonstrated an effective role of computer-guided consultation for screening management. Electronic based systems facilitate real-time decision making during a clinical interaction. Electronic pathways have been applied for clinical and genetic decision support, workflow management, update recommendation and resource estimates. A suitable technical and clinical infrastructure is an integral part of clinical practice guidline of screening. As a conclusion, it is recommended to consider the necessity of architecture assessment and also integration standards.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.6
• /
• pp.89-101
• /
• 2014

In this paper, we proposed a new local micro pattern, Signed Local Directional Pattern(SLDP). SLDP uses information of edges to represent the face's texture. This can produce a more discriminating and efficient code than other state-of-the-art methods. Each micro pattern of SLDP is encoded by sign and its major directions in which maximum edge responses exist-which allows it to distinguish among similar edge patterns that have different intensity transitions. In this paper, we divide the face image into several regions, each of which is used to calculate the distributions of the SLDP codes. Each distribution represents features of the region and these features are concatenated into a feature vector. We carried out facial expression recognition with feature vectors and SVM(Support Vector Machine) on Cohn-Kanade and JAFFE databases. SLDP shows better classification accuracy than other existing methods.

• Journal of Sasang Constitutional Medicine
• /
• v.24 no.2
• /
• pp.1-7
• /
• 2012

1. Objectives : In Sasang constitutional medicine, it has been known that diagnosing Sasang constitution correctly is mostly important. We had developed an Integrated Sasang Constitutional Analysis Tool (SCAT) ${\beta}$-version using face, voice, body shape and questionnaire before. The purpose of this study is to suggest whether SCAT ${\beta}$-version is reliable or not. 2. Methods : We collected 371 subjects from 6 oriental medical clinics. We analyzed the Sasang constitutional diagnostic results using Kappa and coincidence rates between experts in Sasang constitution and SCAT ${\beta}$-version which was developed on the basis of face, body shape, voice and characteristics and symptom questionnaire data. 3. Results : The agreement rates between SCAT ${\beta}$-version and experts was 69.3% in total, and 73.2%, in Taeeumin, 70.8% in Soeumin, and 56.9% in Soyangin in detailed. The Kappa was 0.510 (p value<.000). There was an increasing trend of agreement rates and kappa value corresponded to increasing constitutional probability. When The constitutional probabilities were changed from below 40%, to over 40%, 50%, 60%, the agreement rates corresponded from 50.8% to 79.5%, 91.4%, 95.7% respectively. 4. Conclusions : A SCAT combined with a constitutional probability seemed to help experts to diagnose a patient's Sasang constitution correctly.

• Korean Journal of Organic Agriculture
• /
• v.11 no.2
• /
• pp.23-42
• /
• 2003

The study must have standpoints for the stable market construction of the Environmental-Friendly agricultural products of the quality-certificated which has rapidly grown due to a discussion of the environment and agriculture, income increase and the interests on foods stability. The survey was conducted through face-to-face interview of 200 adults who are in their twenties or more in Cheonan-city. In this research, the Environmental-Friendly agriculture was a clean agriculture not using a fertilizer or chemicals and the agriculture which protects the environment by preventing the environmental pollution. In the analyses of the consumer's willingness to pay, the rice showed 69,851 won, and a lettuce and bean-curd showed 947 won and 1.412 won respectively. In terms of current issues of the policy, to establish the stable circulation structure and consumption strategy, there must be a clearness raise of the Quality Authentication (QA) Mark. To raise the trust through quality authentication, there must be transparency raise of information by distribution stages and the thorough post management of the official institutes. Also, to persue the competitive product differentiation, there must be the settlement of the product brand on the market, development of the new production technology and a classification of consumers by incomes. Finally to construct stable distribution and price system, there must be active participation of the local agricultural cooperatives in the distribution of the Environmental-Friendly agricultural products of the quality-certificated and the understanding of the proper price of the consumer market and flexible strategy of the price change.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.22 no.1
• /
• pp.194-198
• /
• 2008

The treatment before growing into disease is the highest level of the spirit of oriental preventive public medicine. Specially the treatmentt is very important in the life style disease of the aged society, and it is closely connected the human body-form diagnosis reflecting the character of individual physiology and pathology. The human body-form diagnosis explain systematically the color of face, the form and movement of the body, the smell and the taste, the mentality, many or few of $Yin{\cdot}Yang$, the state of pulse, the Meridian, the state of pathology and the direction of the treatment, and therefore the human body-form diagnostis method has also the practical ity on classifying the type. As the spirit of oriental preventive public medicine is based on the Meridian control method and food cooking control method, it has an important effect on the general management of the disease, the making on the beautiful face and the healthy body, the prolongation of life, the management of the individual-ordered health.

• Journal of the Korean association of regional geographers
• /
• v.18 no.2
• /
• pp.129-140
• /
• 2012

In order to examine geomorphic characteristics such as form and retreat process of waterfalls in Jeju Island, 26 waterfalls were classified based upon materials forming a fall face. The waterfalls could be categorized into three types such as single unit, multiple units and basal soft-rock based upon the number and type of a lava flow. A block fall, composed of a pile of large blocks, could be added to the typology of waterfalls in Jeju Island. While the single unit fall is distributed in a region of trachyte or trachyandecite, the multiple units fall and basal soft-rock fall are largely developed in a region of basalt or trachybasalt. A retreat process of the single unit fall is an inclination or a replacement, but the basal soft-rock fall shows a parallel retreat. The multiple units fall exhibits three types of a retreat process according to the physical properties of a lava flow. The fall face is generally vertical due to well-developed vertical joints of a lava flow.

• The Journal of Engineering Geology
• /
• v.17 no.4
• /
• pp.555-566
• /
• 2007

In general, RMR classification system is used for the support design of a tunnel. Face mapping during excavation and RMR-based rock classifications are conducted in order to provide information for complementary changes to preliminary survey plans and for continuous geological estimations in direction of tunnel route. Although they are ever so important, there are not enough time for survey in general and sometimes even face mapping is not available. Linear regression analysis for the estimation of mediating RQD and condition of discontinuities, which require longer time and more detailed observation in RMR, was performed and optimum regression equations are suggest as the result. The geological data collected from tunnels were analyzed in accordance with three rock types as sedimentary rock, phyllite and granite to see geological effects, generally not been considered in previous researches. Parameters for the regression analysis were set another RMR factor.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.2
• /
• pp.78-88
• /
• 2009

In many face recognition problems, the number of available images is limited compared to the dimension of the input space which is usually equal to the number of pixels. This problem is called as the 'small sample size' problem and regularization methods are typically used to solve this problem in feature extraction methods such as LDA. By using regularization methods, the modified within class matrix becomes nonsingu1ar and LDA can be performed in its original form. However, in the process of adding a scaled version of the identity matrix to the original within scatter matrix, the scale factor should be set heuristically and the performance of the recognition system depends on highly the value of the scalar factor. By using the proposed resampling method, we can generate a set of images similar to but slightly different from the original image. With the increased number of images, the small sample size problem is alleviated and the classification performance increases. Unlike regularization method, the resampling method does not suffer from the heuristic setting of the parameter producing better performance.