• The Journal of Society for e-Business Studies
• /
• v.22 no.1
• /
• pp.65-87
• /
• 2017

We propose a method to analyze the user reviews and ratings of the products in the online shopping mall and automatically extracts the features of the products to determine the characteristics of a product. By judging whether a rating is given by a specific feature of a product, our method distributes the score to each feature. Conventional methods force users to wastes time reading overflowing number of reviews and ratings to decide whether to buy the product or not. Moreover, it is difficult to grasp the merits and demerits of the product, because of the way reviews and ratings are provided. It is structured in a way that it is impossible to decide which rating is given to the which characteristics of the product. Therefore, in this paper, to resolve this problem, we propose a method to automatically extract the feature of the product from the user review and distribute the score to appropriate characteristics of the product by calculating the rating of each feature from the overall rating. proposed method collects product reviews and ratings, conducts morphological analysis, and extracts features and emotional words of the products. In addition, a method for determining the polarity of a sentence in which the feature appears is given a weight value for each feature. results of the experiment and the questionnaires comparing the existing methods show the usefulness of the proposed method. We also validates the results by comparing the analysis conducted by the product review experts.

• Applied Biological Chemistry
• /
• v.21 no.3
• /
• pp.193-196
• /
• 1978

Wheat quality score index was proposed as a new criterion of wheat_quality for evaluation of the laboratory testing value such as milling rate (Xm), flour ash content (Xa), flour protein content (Xp) and sedimentation value (Xs). 1) Wheat quality score index was expressed as follows; Milling score index (MSI) $=2.3X_m-72.4X_a$ Protein score index (PSI) =$5.2X_p+0.9X_{psi}$ Composite quality score index (QSI) =$0.63X_{msi}+0.37X_{psi}$ 2) The samples with the more than 100 of MSI would be the good milliability, and the samples with more than 100 of PSI would be the good bread making properties, while that with less than 100 of PSI be good pastry type properties. Whereas most of the samples with higher value of QSI would be related to the good milliability and good bread making properties. 3) MSI, PSI and QSI were significantly correlated to the each related quality components.

• The Korean Journal of Applied Statistics
• /
• v.18 no.3
• /
• pp.543-553
• /
• 2005

Linear logistic regression is one of the most widely used method for credit scoring in credit risk management. This paper deals with credit scoring using splines based on Logistic regression. Linear splines and an automatic basis selection algorithm are adopted. The final model is an example of the generalized additive model. A simulation using a real data set is used to illustrate the performance of the spline method.

• The Korean Journal of Financial Management
• /
• v.19 no.2
• /
• pp.111-134
• /
• 2002

본 연구에서는 건설업종에 특화된 신용평가 모형을 개발하여 건설업종에 대한 부도 예측력를 제고하고자 하였다. 건설업은 여타 업종과는 다른 재무적 특성을 지니고 있다. 특히, 재무적 안정성이 취약하고 자산의 대부분이 매출채권, 재고자산으로 구성되어 유동성이 극히 낮은 실정이다. 본 연구는 이러한 건설업종의 특성을 충분히 감안한 신용평가 모형을 개발하고자 한것이다. 신용평가 모형 중 그 현실적 유용성이 높아 많이 이용되어 오던 신용평점 모형을 개발하였다. 총 2,475개 건설업체를 대상으로 모형구조 및 각종 계량지표 및 비계량지표에 대한 분석을 주로 평균차이 검증과 로짓분석에 의거 선정하였다. 그 결과 새로운 신용평점 모형은 매출액 경상이익률, 총 현금흐름 대 차입금 비율 등 9개의 재무지표와 5분류의 비재무지표로 구성되었다. 이 모형을 기존의 신용평점모형과 비교한 결과 신규모형의 변별력이 높은 것으로 나타났다. 본 연구가 제시한 신용평점모형과 그 개발 방법이 향후 금융기관들의 부실을 줄이고 결과적으로 수익성을 개선하는데 일조하리라 기대된다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.04a
• /
• pp.1285-1286
• /
• 2011

MovieLens를 대상으로 하는 추천 시스템에 대한 연구 중 k-NN 추천 방법은 정확도가 비교적 높지만 평점을 예측할 수 없는 상황이 발생할 수 있다. 본 논문에서는 기존 방법의 문제점을 해결한 장르기반 추천 방법 제안하고, 실험을 통하여 제안하는 방법이 모든 영화에 대한 평점의 예측이 가능함을 검증한다.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.1
• /
• pp.17-24
• /
• 2020

In this paper, we present a movie rating prediction architecture based on a convolutional neural network (CNN). Our prediction architecture extends TextCNN, a popular CNN-based architecture for sentence classification, in three aspects. First, character embeddings are utilized to cover many variants of words since reviews are short and not well-written linguistically. Second, the attention mechanism (i.e., squeeze-and-excitation) is adopted to focus on important features. Third, a scoring function is proposed to convert the output of an activation function to a review score in a certain range (1-10). We evaluated our prediction architecture on a movie review dataset and achieved a low MSE (e.g., 3.3841) compared with an existing method. It showed the superiority of our movie rating prediction architecture.

• Proceedings of the Korean Statistical Society Conference
• /
• 2000.11a
• /
• pp.231-234
• /
• 2000

본 연구는 오랜 시간에 거쳐 축적된 고객 데이터베이스를 활용하여 스코어링 방법을 적용할 수 있는 모델링의 개발에 목적이 있다. 기존의 전통적인 스코어링 방법은 인구 통계학적인 변수나 거래 관련 횡단면적인 자료를 이용하여 우량고객과 불량고객을 구분하는 판별분석의 형태가 대부분이다. 하지만 과거 고객에 대한 실적 자료가 시계열 형태를 이루며 존재하기 때문에 이에 대한 적절한 동태적 모형을 적용은 자연스러운 확장이라고 볼 수 있다. 본 연구에서 제안하는 모형은 고객들의 실적관련 시계열 자료를 GARCH 모형에 적합하여 미래의 실적 예측과 이에 대한 표준편차를 예측하여 하위 $10\%$에 해당하는 실적 예측치를 스코어링으로 하는 새로운 방법을 소개하고자 한다. 이 경우 스코어 값이 부호를 가지게 되므로 우량고객을 구분함과 동시에 큰 음수 값을 조사하여 위험 평점도 함께 측정할 수 있어서 실무 측면에서 유용하리라고 본다.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.6
• /
• pp.81-89
• /
• 2020

Online reviews we encounter commonly on SNS, although a complex range of assessment information affecting the consumer's preferences are included, it is general that such information is just provided by simple numbers or star ratings. Based on those review types, it is not easy to get specific information that consumers want and use it to make a decision for purchase. Therefore, in this study, we propose a prediction methodology that can provide ratings broken down by evaluation items by performing sentiment analysis on restaurant reviews written in Korean. To this end, we select 'food', 'price', 'service', and 'atmosphere' as the main evaluation items of restaurants, and build a new sentiment dictionary for each evaluation item. It also classifies review sentences by rating item, predicts granular ratings through sentiment analysis, and provides additional information that consumers can use to make decisions. Finally, using MAE and RMSE as evaluation indicators it shows that the rating prediction accuracy of the proposed methodology has been improved than previous studies and presents the use case of proposed methodology.

• Journal of KIISE
• /
• v.44 no.12
• /
• pp.1333-1337
• /
• 2017

Recommender systems (RS) that predict a set of items a target user is likely to prefer have been extensively studied in academia and have been aggressively implemented by many companies such as Google, Netflix, eBay, and Amazon. Data imputation alleviates the data sparsity problem occurring in recommender systems by inferring missing ratings and adding them to the original data. In this paper, we point out the drawbacks of existing approaches and make suggestions for data imputation techniques. We also justify our suggestions through extensive experiments.

• The Journal of the Korea Contents Association
• /
• v.21 no.11
• /
• pp.30-40
• /
• 2021

Ratings and reviews, despite their growing influence on restaurants' sales and reputation, entail a few limitations due to the burgeoning of reviews and inaccuracies in rating systems. This study explores the texts in reviews and ratings of a delivery application and discovers ways to elevate review credibility and usefulness. Through a text mining method, we concluded that the delivery application 'Yogiyo' has (1) a five-star oriented rating dispersion, (2) a strong positive correlation between rating factors (taste, quantity, and delivery) and (3) distinct part of speech and morpheme proportions depending on review polarity. We created a chicken-specialized negative word dictionary under four main topics and 20 sub-topic classifications after extracting a total of 367 negative words. We provide insights on how the research on delivery app reviews should progress, centered on fried chicken reviews.