• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.3
• /
• pp.9-15
• /
• 2019

Large global online shopping malls, such as Amazon, offer services in English or in the language of a country when their products are sold. Since many customers purchase products based on the product reviews, the shopping malls actively utilize the sentimental analysis technique in judging preference of each product using the large amount of review data that the customer has written. And the result of such analysis can be used for the marketing to look the potential shoppers. However, it is difficult to apply this English-based semantic analysis system to different languages used around the world. In this study, more than 500,000 data from Amazon fine food reviews was used for training a deep learning based system. First, sentiment analysis evaluation experiments were carried out with three models of English test data. Secondly, the same data was translated into seven languages (Korean, Japanese, Chinese, Vietnamese, French, German and English) and then the similar experiments were done. The result suggests that although the accuracy of the sentimental analysis was 2.77% lower than the average of the seven countries (91.59%) compared to the English (94.35%), it is believed that the results of the experiment can be used for practical applications.

• Proceedings of the Korea Contents Association Conference
• /
• 2018.05a
• /
• pp.19-20
• /
• 2018

전자상거래가 대중화되면서 다양한 아이템을 손쉽게 구매할 수 있는 환경이 조성되었다. 전자상거래에서 소비자의 구매율을 향상시키기 위해 개인 맞춤 추천 서비스가 요구되고 있다. 본 논문에서는 사용자 성향과 제품의 신뢰성을 고려한 상품 추천 기법을 제안한다. 사용자의 성향은 찜하기, 리뷰, 클릭 등과 같은 다양한 사용자의 행위 분석을 통해 추출하고 상품의 신뢰성은 SNS에서의 언급 수와 서비스내의 사용자 행위를 통해 계산한다. 계산된 성향을 기반으로 협업 필터링을 수행하여 상품별 예측 점수를 생성하고 상품의 신뢰성을 고려하여 최종적인 추천 목록을 생성한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2017.01a
• /
• pp.257-259
• /
• 2017

본 논문은 여행상품 일정의 POI정보를 기반으로 생성된 여행 스케줄러에 따라 실제 여행이 이루어지고 여행 중에 촬영된 사진과 여행자가 작성한 여행상품 리뷰 및 여행기 등의 정보를 매시업하여 여행노트를 생성하는 시스템을 구현하였다. 무엇보다 여행자가 일일이 자신의 여행 스케줄을 입력해야하는 번거로움을 없이 여행중에 편리성을 제공받을 수 있다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.502-503
• /
• 2019

In the existing system that recommends through review data analysis, it does not reflect personal preference details such as user's characteristics or product purchase tastes, in this paper, we propose a system that provides customized recommendation information to various users by selecting the criterion that the user thinks most importantly when searching for the product and purchasing the product, and analyzing it. This is because the user's personal preference is reflected by arranging the product list based on the criterion that the user occupies the largest portion of the product purchase, so that it is more efficient than the recommendation through the recommendation system.

• Information Systems Review
• /
• v.25 no.1
• /
• pp.1-26
• /
• 2023

Consumers tend to trust information provided by other consumers more than information provided by sellers. Therefore, while inducing consumers to write high-quality reviews is a very important task for companies, it is not easy to produce such high-quality reviews. Based on previous research on review writing and memory recall, we decided to develop a way to use digital nudge to help consumers naturally write high-quality reviews. Specifically, we designed an experiment to verify the effect of uploading a photo during the online review process on the quality of review of the review writer. We then recruited subjects and then divided them into groups that upload photos first and groups that do not. A task was assigned to each subject to write positive and negative reviews. As a result, it was confirmed that the behavior of uploading a photo first increases the review length. In addition, it was confirmed that when online users who upload photos first have extremely negative satisfaction with the product, the extent of two-sidedness of the review content increases.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.501-503
• /
• 2012

오피니언 마이닝에서 특징기반으로 상품평을 요약할 때, 동일한 상품의 같은 특징에 대한 사용자의 표현이 일치하지 않아 같은 특징을 다른 것으로 인식하는 오류가 발생되어 효과적인 분석을 하는데 어려움이 있다. 본 연구에서는 이러한 문제점을 해결하기 위하여 온라인쇼핑몰의 상품평에서 명사와 형용사쌍 말뭉치를 이용하여 연관단어뭉치를 추출하고, 상관성이 높은 형용사를 각 명사의 특징으로 이용하여 대체어 목록을 자동으로 추출하는 방법을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.11a
• /
• pp.35-38
• /
• 2010

인터넷 시장이 빠르게 성장함에 따라 사용자들의 참여도가 매우 높아졌다. 인터넷 사용자들은 인터넷 쇼핑의 상품에 관한 의견을 웹 상에 표현하기 시작했고, 실제 소비자이 판단하는 데에 많은 영향을 미치고 있다. 하지만 현재에 들어 그 양이 엄청나게 방대해 졌기 때문에 사용자들이 원하는 정보만을 찾아내는 것은 어려운 일이다. 본 논문에서는 사용들이 작성한 인터넷 쇼핑에서 상품평에 관한 리뷰를 모아 방대한 양에서 오피니언 마이닝 기법을 이용해 유용한 정보를 효율적으로 도출해서 사용자가 원하는 정보를 요약하여 제공하는 방법을 제안한다. 이러한 방법을 통해서 사용자는 상품을 구매하기 전에 좀 더 객관적이고 효율적으로 판단을 내릴 수 있을 것이다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.759-760
• /
• 2020

최근 사용자, 상품, 그리고 상품의 메타 정보 사이의 관계를 표현한 지식 그래프 (knowledge graph) 가 추천 시스템 분야에서 많은 관심을 받고 있으며 활발히 이용되고 있다. 하지만 기존의 지식 그래프는 각 노드 (사용자, 상품, 메타 정보 등) 사이의 단순한 사실 관계만을 표현하고 있으며, 이는 사용자의 선호도를 정확히 파악하는 데 한계가 있다. 본 논문에서는 지식 그래프의 정보 부족 문제를 보완하기 위해 각 상품에 남겨진 텍스트 리뷰를 감정 분석 (sentiment analysis) 하고, 이를 각 노드 간의 선호도 정보로 활용하여 지식 그래프를 구축하는 방법을 제안한다.

• Informatization Policy
• /
• v.20 no.3
• /
• pp.3-17
• /
• 2013

As electronic marketplaces grow and a large number of consumers exchange their opinions on products and services on the Internet, many studies have been conducted in the area of online consumer reviews. This paper analyzes the research trend of the online consumer reviews by investigating those studies in an attempt to provide future research directions. Many researchers have focused on the effects of online reviews on consumer behaviors as well as the usefulness of the online reviews. In particular, review contents, characteristics of reviewers/consumers and features of products/services have been identified as influencing factors on the effects of the online consumer reviews. For the review contents, the number and the volume of the contents have increasing effects on the online reviews, while the direction (positive vs. negative) of the contents has resulted in conflicting effects of the review. The reputation and trustfulness of reviewers, consumers' prior knowledge on the products, consumers' product involvement, and types of the products were investigated as these factors influence the effectiveness of the online consumer reviews. Social media (such as Facebook and Twitter) nowadays play an important role to disseminate online reviews among consumers. Thus, it is necessary to study how social media influence the effects of online reviews on consumers. Since some firms abuse the online reviews for their own sakes, we recognize the necessity for empirical studies on the side effects of the online reviews.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.97-117
• /
• 2020

Product evaluation criteria is an indicator describing attributes or values of products, which enable users or manufacturers measure and understand the products. When companies analyze their products or compare them with competitors, appropriate criteria must be selected for objective evaluation. The criteria should show the features of products that consumers considered when they purchased, used and evaluated the products. However, current evaluation criteria do not reflect different consumers' opinion from product to product. Previous studies tried to used online reviews from e-commerce sites that reflect consumer opinions to extract the features and topics of products and use them as evaluation criteria. However, there is still a limit that they produce irrelevant criteria to products due to extracted or improper words are not refined. To overcome this limitation, this research suggests LDA-k-NN model which extracts possible criteria words from online reviews by using LDA and refines them with k-nearest neighbor. Proposed approach starts with preparation phase, which is constructed with 6 steps. At first, it collects review data from e-commerce websites. Most e-commerce websites classify their selling items by high-level, middle-level, and low-level categories. Review data for preparation phase are gathered from each middle-level category and collapsed later, which is to present single high-level category. Next, nouns, adjectives, adverbs, and verbs are extracted from reviews by getting part of speech information using morpheme analysis module. After preprocessing, words per each topic from review are shown with LDA and only nouns in topic words are chosen as potential words for criteria. Then, words are tagged based on possibility of criteria for each middle-level category. Next, every tagged word is vectorized by pre-trained word embedding model. Finally, k-nearest neighbor case-based approach is used to classify each word with tags. After setting up preparation phase, criteria extraction phase is conducted with low-level categories. This phase starts with crawling reviews in the corresponding low-level category. Same preprocessing as preparation phase is conducted using morpheme analysis module and LDA. Possible criteria words are extracted by getting nouns from the data and vectorized by pre-trained word embedding model. Finally, evaluation criteria are extracted by refining possible criteria words using k-nearest neighbor approach and reference proportion of each word in the words set. To evaluate the performance of the proposed model, an experiment was conducted with review on '11st', one of the biggest e-commerce companies in Korea. Review data were from 'Electronics/Digital' section, one of high-level categories in 11st. For performance evaluation of suggested model, three other models were used for comparing with the suggested model; actual criteria of 11st, a model that extracts nouns by morpheme analysis module and refines them according to word frequency, and a model that extracts nouns from LDA topics and refines them by word frequency. The performance evaluation was set to predict evaluation criteria of 10 low-level categories with the suggested model and 3 models above. Criteria words extracted from each model were combined into a single words set and it was used for survey questionnaires. In the survey, respondents chose every item they consider as appropriate criteria for each category. Each model got its score when chosen words were extracted from that model. The suggested model had higher scores than other models in 8 out of 10 low-level categories. By conducting paired t-tests on scores of each model, we confirmed that the suggested model shows better performance in 26 tests out of 30. In addition, the suggested model was the best model in terms of accuracy. This research proposes evaluation criteria extracting method that combines topic extraction using LDA and refinement with k-nearest neighbor approach. This method overcomes the limits of previous dictionary-based models and frequency-based refinement models. This study can contribute to improve review analysis for deriving business insights in e-commerce market.