• Journal of KIISE:Software and Applications
• /
• v.37 no.6
• /
• pp.470-479
• /
• 2010

In relation to the extraction of opinion words, it may be difficult to directly apply most of the methods suggested in existing English studies to the Korean language. Additionally, the manual method suggested by studies in Korea poses a problem with the extraction of opinion words in that it takes a long time. In addition, English thesaurus-based extraction of Korean opinion words leaves a challenge to reconsider the deterioration of precision attributed to the one to one mismatching between Korean and English words. Studies based on Korean phrase analyzers may potentially fail due to the fact that they select opinion words with a low level of frequency. Therefore, this study will suggest the k-Structure (k=5 or 8) method, which may possibly improve the precision while mutually complementing existing studies in Korea, in automatically extracting opinion words from a simple sentence in a given Korean product review. A simple sentence is defined to be composed of at least 3 words, i.e., a sentence including an opinion word in ${\pm}2$ distance from the attribute name (e.g., the 'battery' of a camera) of a evaluated product (e.g., a 'camera'). In the performance experiment, the precision of those opinion words for 8 previously given attribute names were automatically extracted and estimated for 1,868 product reviews collected from major domestic shopping malls, by using k-Structure. The results showed that k=5 led to a recall of 79.0% and a precision of 87.0%; while k=8 led to a recall of 92.35% and a precision of 89.3%. Also, a test was conducted using PMI-IR (Pointwise Mutual Information - Information Retrieval) out of those methods suggested in English studies, which resulted in a recall of 55% and a precision of 57%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.10
• /
• pp.2818-2839
• /
• 2024

With the continuous growth of e-commerce sales, a robust product selection model is essential to maintain competitiveness and meet consumer demand. Current research primarily focuses on single models for sales prediction and lacks an integrated approach to sales forecasting and product selection. This paper proposes a comprehensive framework (VN-CPC) that combines sales forecasting with product selection to address these issues. We integrate a series of classical machine learning models, including Tree Models (XGBoost, LightGBM, CatBoost), Support Vector Machine (SVM), Bayesian Ridge, and Artificial Neural Networks (ANN), using a voting mechanism to determine the optimal weighting scheme. Our method demonstrates a lower Root Mean Square Error (RMSE) on collected Amazon data than individual models and other ensemble models. Furthermore, we employ a three-tiered clustering model: Initial Clustering, Refinement Clustering, and Final Clustering, based on our predictive model to refine product selection to specific categories. This integrated forecasting and selection framework can be more effectively applied in the dynamic e-commerce environment. It provides a robust tool for businesses to optimize their product offerings and stay ahead in a competitive market.

• Annual Conference of KIPS
• /
• 2008.05a
• /
• pp.107-110
• /
• 2008

상품평은 인터넷 쇼핑 이용자들의 최종 구매결정에 큰 영향을 미치는 것으로 알려져 있다. 많은 쇼핑몰에서 상품평 활성화를 위해 노력하고 있지만, 상품평을 모으는 것에만 주력할 뿐 기존에 수집된 상품평을 제공하는 방법에 있어서는 원시적인 수준에 그치고 있다. 상품평을 좀 더 효율적으로 제공하려면 사용자들이 상품평에서 찾게 될 평가항목들을 미리 예측하여 그 항목에 따라 상품평을 분류/요약해서 제공하는 방법을 생각할 수 있다. 본 논문에서는 상품평과 웹 검색엔진을 이용하여 각 상품별 평가항목들을 자동으로 추출하는 방법을 제안한다. 상품평 데이터의 특성상 노이즈가 많기 때문에 먼저 데이터를 정제하고, 정제된 상품평 데이터를 형태소 분석하여 후보명사들을 선택한다. 선택된 후보명사를 웹 검색엔진에 질의하여 반환된 결과 값으로 상품 카테고리와 후보명사 간 연관도를 계산하여 평가항목을 추출한다. 실험은 5개 상품 카테고리의 170,294개 실제 상품평을 대상으로 각 카테고리별 평가항목을 추출하였다.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.4
• /
• pp.142-151
• /
• 2023

Sentiment analysis is a method used to comprehend feelings, opinions, and attitudes in text, and it is essential for evaluating consumer feedback and social media posts. However, creating sentiment dictionaries, which are necessary for this analysis, is complex and time-consuming because people express their emotions differently depending on the context and domain. In this study, we propose a new method for simplifying this procedure. We utilize syntax analysis of the Korean language to identify and extract sentiment words based on the Reason-Sentiment Pattern, which distinguishes between words expressing feelings and words explaining why those feelings are expressed, making it applicable in various contexts and domains. We also define sentiment words as those with clear polarity, even when used independently and exclude words whose polarity varies with context and domain. This approach enables the extraction of explicit sentiment expressions, enhancing the accuracy of sentiment analysis at the attribute level. Our methodology, validated using Korean cosmetics review datasets from Korean online shopping malls, demonstrates how a sentiment dictionary focused solely on clear polarity words can provide valuable insights for product planners. Understanding the polarity and reasons behind specific attributes enables improvement of product weaknesses and emphasis on strengths. This approach not only reduces dependency on extensive sentiment dictionaries but also offers high accuracy and applicability across various domains.

• The Journal of the Korea Contents Association
• /
• v.9 no.6
• /
• pp.126-133
• /
• 2009

This study examined the relationship of what kind of creative thinking has as factors for emotion design products for consumers focusing on the successful cases of emotion products. For the design creativity attribute used in this experiment, the design evaluation creativity tools revealed in Kim Eun-Ju's 2007 design creativity evaluation tool development were used mostly MP3s, which have various forms, functions and sizes were selected as the target for experiment. Results of the experiment showed that for design creativeness items for MP3 as single products, uniqueness, favorableness and convenience were relevant. Accordingly, the common features of design creativeness items for emotional products were identified. According to the result, for emotional designs, the interest level of uniqueness for the design creativity evaluation items and the functional items for practicality had a high relativity. Therefore, there is a need to examine the common features between the design creativity items for products other than MP3s in the future.