• 제목/요약/키워드: Evaluation Set

검색결과 3,265건 처리시간 0.028초

협업필터링에서 고객의 평가치를 이용한 선호도 예측의 사전평가에 관한 연구 (Pre-Evaluation for Prediction Accuracy by Using the Customer's Ratings in Collaborative Filtering)

  • 이석준;김선옥
    • Asia pacific journal of information systems
    • /
    • 제17권4호
    • /
    • pp.187-206
    • /
    • 2007
  • The development of computer and information technology has been combined with the information superhighway internet infrastructure, so information widely spreads not only in special fields but also in the daily lives of people. Information ubiquity influences the traditional way of transaction, and leads a new E-commerce which distinguishes from the existing E-commerce. Not only goods as physical but also service as non-physical come into E-commerce. As the scale of E-Commerce is being enlarged as well. It keeps people from finding information they want. Recommender systems are now becoming the main tools for E-Commerce to mitigate the information overload. Recommender systems can be defined as systems for suggesting some Items(goods or service) considering customers' interests or tastes. They are being used by E-commerce web sites to suggest products to their customers who want to find something for them and to provide them with information to help them decide which to purchase. There are several approaches of recommending goods to customer in recommender system but in this study, the main subject is focused on collaborative filtering technique. This study presents a possibility of pre-evaluation for the prediction performance of customer's preference in collaborative filtering before the process of customer's preference prediction. Pre-evaluation for the prediction performance of each customer having low performance is classified by using the statistical features of ratings rated by each customer is conducted before the prediction process. In this study, MovieLens 100K dataset is used to analyze the accuracy of classification. The classification criteria are set by using the training sets divided 80% from the 100K dataset. In the process of classification, the customers are divided into two groups, classified group and non classified group. To compare the prediction performance of classified group and non classified group, the prediction process runs the 20% test set through the Neighborhood Based Collaborative Filtering Algorithm and Correspondence Mean Algorithm. The prediction errors from those prediction algorithm are allocated to each customer and compared with each user's error. Research hypothesis : Two research hypotheses are formulated in this study to test the accuracy of the classification criterion as follows. Hypothesis 1: The estimation accuracy of groups classified according to the standard deviation of each user's ratings has significant difference. To test the Hypothesis 1, the standard deviation is calculated for each user in training set which is divided 80% from MovieLens 100K dataset. Four groups are classified according to the quartile of the each user's standard deviations. It is compared to test the estimation errors of each group which results from test set are significantly different. Hypothesis 2: The estimation accuracy of groups that are classified according to the distribution of each user's ratings have significant differences. To test the Hypothesis 2, the distributions of each user's ratings are compared with the distribution of ratings of all customers in training set which is divided 80% from MovieLens 100K dataset. It assumes that the customers whose ratings' distribution are different from that of all customers would have low performance, so six types of different distributions are set to be compared. The test groups are classified into fit group or non-fit group according to the each type of different distribution assumed. The degrees in accordance with each type of distribution and each customer's distributions are tested by the test of ${\chi}^2$ goodness-of-fit and classified two groups for testing the difference of the mean of errors. Also, the degree of goodness-of-fit with the distribution of each user's ratings and the average distribution of the ratings in the training set are closely related to the prediction errors from those prediction algorithms. Through this study, the customers who have lower performance of prediction than the rest in the system are classified by those two criteria, which are set by statistical features of customers ratings in the training set, before the prediction process.

문화콘텐츠 가치평가의 평가모형에 관한 연구 (A Study of Evaluation Model for Culture Contents' Value Evaluation)

  • 권지혁;백승국;손기동
    • 디지털산업정보학회논문지
    • /
    • 제9권3호
    • /
    • pp.129-144
    • /
    • 2013
  • Recognizing a limits on quantitative evaluation of cultural contents' and for its betterment, study aims developing a qualitative evaluation model. For this study, Reception Theory, Semiotics and Psychology were derived for epistemological dimension to contemplate culture contents' essential attribute. To be concrete, cultural contents was examined as experiential products, emotional products, rememberable products and texts. Also, codes of fun, emotion and culture were discussed as intrinsic attributes for cultural contents and how those attributes were expressed or composed in cultural contents was discussed as well. Evaluation items were extracted based on final discussion at the epistemic level, set up the final evaluation model by taking experts' advices on each items. With all those outcomes, qualitative evaluation model for cultural contents was developed. For the importance of each index in the model, priority was granted by weighting on each index. Lastly, evaluation scale was developed for each index. The culture contents' evaluation model developed in study is meaningful not only in drawing qualitative evaluation items of video(image) contents and developing the index and model for the first time, but also its possibility of wide use for other genres.

원자력 증기용 안전밸브의 개방성능 평가를 위한 해석적 연구 (An Analytical Study on Evaluation of Opening Performance of Steam Safety Valve for Nuclear Power Plant)

  • 손상호
    • 한국유체기계학회 논문집
    • /
    • 제17권1호
    • /
    • pp.5-11
    • /
    • 2014
  • The purpose of this paper is to investigate an analytical approach for opening performance evaluation of the nuclear pressure safety valve based on standard codes such as ASME or KEPIC. It is well-known that safety valve is considered as one of pressure relief valves for protecting a boiler or pressure vessel from exceeding the maximum allowable working pressure. When pressure in a container reaches its set pressure, the safety valve commences discharging the internal fluid by a sudden opening called as popping. Safety valve is usually evaluated by set pressure, full open, blow-down, leakage and flow capacity. The test procedure and technical requirement for performance evaluation is described in international code of ASME code such as BPVC. The opening characteristics of steam safety valve can be analyzed by computational fluid dynamics (CFD) and steam shaft dynamics. First, the flow analysis along opening process is simulated by running the CFD models of the ten types of opening steps from 0 to 100%. As a analysis result, the various CFD outputs of flow pattern, pressure, forces on the disc and mass flow at each simulation step is demonstrated. The lift force is calculated by using the forces applied on disc from static pressure and secondary flow. And, the effect of huddle chamber or control chamber is studied by dynamic analysis based on CFD simulation results such as lift force. As a result, dynamics analysis shows opening features according to the sizes of control chamber.

분전반 관리시스템 평가를 위한 시험 장치의 제작 및 특성 분석 (Manufacturing and Characteristics Analysis of a Testing Device for the Evaluation of a Distribution Board Management System)

  • 고완수;이병설;최충석
    • 한국안전학회지
    • /
    • 제34권5호
    • /
    • pp.31-36
    • /
    • 2019
  • This study made a testing device to evaluate the distribution board management system. Power was supplied to the testing device using a loading-back method and the voltage applied to it was 440 V at the same turn ratio. When the human body electric shock current is 30 mA, the breaking time is set to be less than 240 ms while 30~45 mA current is flowing. The test result shows that in the case of the R-phase it was measured to be 5.19 Hz (193 ms). And the S-phase and T-phase were perfectly cut off at 5.39 Hz (186 ms) and 5.71 Hz (175 ms), respectively. When the human body electric shock current is 60mA, the breaking time is set to be less than 120 ms while 45~75 mA current is flowing. The test result shows that the R-phase, S-phase, and T-phase were accurately cut off at 8.39 Hz (11 ms), 8.87Hz (113 ms) and 9.69 Hz (103 ms), respectively. When the human body electric shock current is 90 mA, the breaking time is set to be less than 48 ms while 75 mA current is flowing. The test result shows that the R-phase, S-phase, and T-phase were accurately cut off at 19.8 Hz (50.4 ms), 16.9 Hz (59.2 ms), and 17.9 Hz (56.0 ms), respectively. That is, the developed testing device satisfied all the requirements of the distribution board evaluation criteria, and it becomes available for the performance evaluation of the distribution board management system.

잡음 환경에서의 음성인식을 위한 온라인 빔포밍과 스펙트럼 감산의 결합 (Combining deep learning-based online beamforming with spectral subtraction for speech recognition in noisy environments)

  • 윤성욱;권오욱
    • 한국음향학회지
    • /
    • 제40권5호
    • /
    • pp.439-451
    • /
    • 2021
  • 본 논문에서는 실제 환경에서의 연속 음성 강화를 위한 딥러닝 기반 온라인 빔포밍 알고리듬과 스펙트럼 감산을 결합한 빔포머를 제안한다. 기존 빔포밍 시스템은 컴퓨터에서 음성과 잡음을 완전히 겹친 방식으로 혼합하여 생성된 사전 분할 오디오 신호를 사용하여 대부분 평가되었다. 하지만 실제 환경에서는 시간 축으로 음성 발화가 띄엄띄엄 발성되기 때문에, 음성이 없는 잡음 신호가 시스템에 입력되면 기존 빔포밍 알고리듬의 성능이 저하된다. 이러한 효과를 경감하기 위하여, 심층 학습 기반 온라인 빔포밍 알고리듬과 스펙트럼 감산을 결합하였다. 잡음 환경에서 온라인 빔포밍 알고리듬을 평가하기 위해 연속 음성 강화 세트를 구성하였다. 평가 세트는 CHiME3 평가 세트에서 추출한 음성 발화와 CHiME3 배경 잡음 및 MUSDB에서 추출한 연속 재생되는 배경음악을 혼합하여 구성되었다. 음성인식기로는 Kaldi 기반 툴킷 및 구글 웹 음성인식기를 사용하였다. 제안한 온라인 빔포밍 알고리듬 과 스펙트럼 감산이 베이스라인 빔포밍 알고리듬에 비해 성능 향상을 보임을 확인하였다.

슬관절 전치환술 후 한의 핵심 결과 지표를 개발하기 위한 임상 평가지표에 대한 문헌 연구 (A Literature Study about Clinical Outcome Parameters for Total Knee Replacement to Develop Core Outcome Set for Osteoarthritis by Korean Medicine Treatment)

  • 전채헌;김혜진;이정민;권미리;장승원;김현호;공병희;임정태
    • 한방재활의학과학회지
    • /
    • 제29권3호
    • /
    • pp.51-60
    • /
    • 2019
  • Objectives Osteoarthritis is hard to manage with both conventional and Korean medicine treatment. The core outcome set (COS) to demonstrate the effectiveness of Korean medicine treatment has not been established yet. We aimed to present preliminary data of COS by performing a literature review on the evaluation indices used in existing clinical research. Methods We examined the literature from 2000 to 2017 in two Korean electronic databases (Korea citation index and oriental medicine advanced searching integrated system) by searching for the following 3 terms 'total knee replacement (Korean)', 'total knee replacement,' and 'knee surgery.' We found 333 articles; among them, 50 duplicates were removed. Finally, we selected 160 articles after complete screening. We then extracted measured indices and clinical outcomes from the selected articles and categorized the relevant criteria. Results According to this study, the hospital for special surgery and knee society, range of movement angle, cross leg, Berg balance scale and balance ability, muscle strength, 6 minutes walking test, visual analogue scale, self-efficacy, the 12-item and 36-item short form survey and self-rated health status are the most commonly used outcomes of knee. Conclusions This study found that the several categories after total knee replacement (TKR) are being evaluated in the literature, and we were able to verify the most frequently used evaluation indices in these categories. The results of this study will be used to establish evaluation indices for the treatment of TKR in the future using Korean medicine.

벼품종의 재배지역에 따른 미질특성변이 II. 미질관련형질 상호간의 관계 (Variation of Grain Quality of Rice Varieties Grown at Different Locations II. Relationship between Characteristics Related to Grain Quality)

  • 김광호;주현규
    • 한국작물학회지
    • /
    • 제35권2호
    • /
    • pp.137-145
    • /
    • 1990
  • 국내에서 재배되고 있는 벼 6품종을 대상으로 품종별로 8-20 지역에서 1987년과 1988년에 생산된 쌀을 각각 수집하여 그들의 외관, 아밀로스함량 및 알칼리붕괴도, 아밀로그라프에 의한 쌀가루의 호화 및 점성특성과 레오메타를 이용한 밥알의 조직감(Texture)을 조사하고 관능검사법에 의한 식미평가를 실시하여 이들 상호간의 관계를 검토하였다. 1. 동일품종내에서는 재배지역에 따른 심복백미비율의 변이가 백미천립중, 쌀알의 아밀로스 함량과 알칼리붕괴도, 쌀가루의 아밀로그람특성, 밥알의 레오그람특성 및 식미와 각각 독립적으로 나타났다. 2. 재배지역에 따른 변이정도가 컸던 형질들간의 단순상관관계를 보면 아밀로그람특성 중 최고점도와 break down간에 정의 상관, 최고점도와 set back간, 그리고 set back과 break down간에는 부의 상관이 인정되었고 알칼리붕괴도와 break down간, 알칼리붕괴도와 최고점도간에는 정의 상관, 그리고 알칼리붕괴도와 set back간에는 부의 상관이 인정되었다. 3. 아밀로그람특성 중 최고점도와 break down 값이 낮았고 동시에 set back 값은 높았던 3지역 쌀은 최고점도와 break down 값이 높고 set back 값이 낮았던 3지역 쌀에 비하여 알칼리붕괴도는 높고, 레오그람특성 중 식미지수(TPI)와 점성/경도비율(Vi/H)은 낮았으며 관능검사법에 의한 종합식미에서도 낮은 점수를 받았다. 4. 레오메타로 조사한 점성/경도비율과 관능검사로 평가한 종합식미를 품종별 지역평균치보다 상급, 하급 그리고 평균치정도의 3등급으로만 나누어 두 형질간의 간계를 검토한 결과 점성/경도비율이 상급인 쌀은 종합식미가 평균치정도 또는 상급으로 판정받은 경우가 대부분이었다. 레오메타특성 중 식미지수와 종합식미와의 관계도 점성/경도비율에서와 비슷한 경향을 보였다.

  • PDF

Development of Evaluation Criteria for Forest Education Using the CIPP Model

  • Kim, Soyeon;Choi, Jungkee
    • Journal of Forest and Environmental Science
    • /
    • 제36권2호
    • /
    • pp.163-172
    • /
    • 2020
  • The objective of this study was to develop evaluation criteria for forest education using the Context, Input, Process, and Product (CIPP) model. To this end, we designed a survey based on expert advice and content analysis of previous studies on the CIPP model and forest education. The survey was conducted on 393 forest education specialists, and Cronbach's α coefficient was set as 0.6 or higher to verify reliability and validity, and to determine reliability by factor. Eventually, 52 out of 57 evaluation items were extracted, and the evaluation indexes were selected through factor analysis as follows: four evaluation indexes for the context dimension, namely "Clarity of goal setting," "Developing conditions for education," "Meeting of requirements," and "Institutional drive"; three evaluation indexes for the input dimension, namely "Acquisition of education infrastructure," "Establishment of operational support," and "Adequacy of assigned manpower"; four evaluation indexes for the process dimension, which were "Adequacy of budget allocation," "Expertise of forest education instructors," "Diversity of programs," and "Public-private academic partnership"; and five evaluation indexes for the product dimension, namely "Effectiveness of perception change," "Influence over the society," "Continuity of improvement in evaluation," "Continuity of education," and "Verification of the effects of education."

Accuracy evaluation of liver and tumor auto-segmentation in CT images using 2D CoordConv DeepLab V3+ model in radiotherapy

  • An, Na young;Kang, Young-nam
    • 대한의용생체공학회:의공학회지
    • /
    • 제43권5호
    • /
    • pp.341-352
    • /
    • 2022
  • Medical image segmentation is the most important task in radiation therapy. Especially, when segmenting medical images, the liver is one of the most difficult organs to segment because it has various shapes and is close to other organs. Therefore, automatic segmentation of the liver in computed tomography (CT) images is a difficult task. Since tumors also have low contrast in surrounding tissues, and the shape, location, size, and number of tumors vary from patient to patient, accurate tumor segmentation takes a long time. In this study, we propose a method algorithm for automatically segmenting the liver and tumor for this purpose. As an advantage of setting the boundaries of the tumor, the liver and tumor were automatically segmented from the CT image using the 2D CoordConv DeepLab V3+ model using the CoordConv layer. For tumors, only cropped liver images were used to improve accuracy. Additionally, to increase the segmentation accuracy, augmentation, preprocess, loss function, and hyperparameter were used to find optimal values. We compared the CoordConv DeepLab v3+ model using the CoordConv layer and the DeepLab V3+ model without the CoordConv layer to determine whether they affected the segmentation accuracy. The data sets used included 131 hepatic tumor segmentation (LiTS) challenge data sets (100 train sets, 16 validation sets, and 15 test sets). Additional learned data were tested using 15 clinical data from Seoul St. Mary's Hospital. The evaluation was compared with the study results learned with a two-dimensional deep learning-based model. Dice values without the CoordConv layer achieved 0.965 ± 0.01 for liver segmentation and 0.925 ± 0.04 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.927 ± 0.02 for liver division and 0.903 ± 0.05 for tumor division. The dice values using the CoordConv layer achieved 0.989 ± 0.02 for liver segmentation and 0.937 ± 0.07 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.944 ± 0.02 for liver division and 0.916 ± 0.18 for tumor division. The use of CoordConv layers improves the segmentation accuracy. The highest of the most recently published values were 0.960 and 0.749 for liver and tumor division, respectively. However, better performance was achieved with 0.989 and 0.937 results for liver and tumor, which would have been used with the algorithm proposed in this study. The algorithm proposed in this study can play a useful role in treatment planning by improving contouring accuracy and reducing time when segmentation evaluation of liver and tumor is performed. And accurate identification of liver anatomy in medical imaging applications, such as surgical planning, as well as radiotherapy, which can leverage the findings of this study, can help clinical evaluation of the risks and benefits of liver intervention.

결막 충혈도 측정을 위한 공막 영상 분할 (Sclera Segmentation for the Measurement of Conjunctival Injection)

  • 배장표;김광기;정창부;양희경;황정민
    • 한국멀티미디어학회논문지
    • /
    • 제13권8호
    • /
    • pp.1142-1153
    • /
    • 2010
  • 결막 충혈은 결막염, 각막염, 포도막염 등의 안과질환의 초기 증세로서 정량적으로 평가할 수 있다면 진단과 경과 관찰에 도움이 된다. 충혈의 정량화에서 공막의 크기는 중요한 지표이지만 기존의 공막 분할 방법이 정확하지 않기 때문에 수동으로 분할하고 있다. 본 논문에서는 충혈의 정량화를 위하여 level set 방법을 이용한 공막 분할 알고리즘을 제안한다. Level set의 초기 모델은 Lab 색상 모드와 k-means 알고리즘, 기하학적인 정보를 이용하여 지정된다. 헤이시안(hessian) 분석으로 공막과 피부 사이의 골을 향상시킨 영상에 level set을 적용하였다. 제안 방법의 성능 측정을 위하여 52개의 전안부 영상에 대하여 실험하였다. 실험 결과, 제안 방법이 화소값만 이용하는 region growing이나 level set의 초기 모델로 임의의 원을 이용하는 방법보다 성능이 우수하였다. 이 논문에서 제안한 공막 분할 방법은 객관적인 충혈도 측정에서 중요한 요소 기술의 역할을 할 것이다.