• 한국벤처창업학회:학술대회논문집
• /
• 2022.11a
• /
• pp.181-185
• /
• 2022

고객들은 조작된 온라인 리뷰가 범람하는 가운데 진정성과 가치를 지닌 리뷰를 보고자한다. 귀인 이론(Attribution theory)의 관점에서, 사람들은 리뷰어의 과거 평가 이력을 바탕으로 리뷰가 진정성 있는지를 판단하는 경향이 있다. 이러한 배경에서 본 연구의 목적은 리뷰어의 과거 평점 이력이 조작된 리뷰로 인식하는 것에 어떠한 영향을 미치며, 최종적으로 리뷰 유용성이 어떠한 영향을 미치는지 알아보는 것이다. 제안된 가설을 검증하기 위해 2차 데이터 분석(연구1)과 실험(연구2)을 수행했으며, 두 연구는 일관된 결과를 보여준다. 연구 1은 리뷰어의 과거 평가 이력이 리뷰 유용성에 미치는 영향을 분석하였다. 귀인이론에 근거하면, 사람들은 리뷰를 다른 목적을 가지고 작성되었다고 인식할 경우에 리뷰가 조작되었다고 생각하고, 그 리뷰가 물건이나 서비스의 진정한 가치를 평가하지 않았다고 간주한다. 따라서 해당 리뷰는 유용성이 낮게 평가되는 경향이 있다. 2차 데이터를 분석하기 위해 우리는 Python을 이용한 웹 스크레이퍼를 개발하여 TripAdvisor(TripAdvisor.com)에서 호텔 정보, 리뷰, 리뷰 정보 등의 연구 데이터를 수집하였다. 수집한 890명 리뷰어에 대한 100,621개의 리뷰를 분석하기 위해 음이항 회귀 분석을 수행하였다. 분석 결과, 평균 평점을 낮게 주는 리뷰어의 경우에 리뷰 유용성에 유의미한 영향을 미치지 않는 것으로 나타났다. 사람들은 극단적인 평점을 거의 주지 않는 리뷰어가 작성한 리뷰가 더 도움이 된다고 평가했다. 연구 2는 리뷰어의 과거 평점 이력을 기준으로 리뷰가 조작되었다고 평가하는 사람들의 인식 프로세스를 실험하였다. 실험 결과, 사람들은 리뷰어의 과거 평점 이력이 평균적으로 평점을 낮게 주는 경우에는 리뷰가 의심스럽다고 판단하지 않는 것으로 나타났다. 그리고 사람들은 리뷰어가 대부분 극단적인 평점을 주는 이력이 있다면 해당 리뷰어가 작성한 리뷰가 의심스럽다고 판단하는 것으로 나타났다. 연구2는 사람들이 리뷰어의 과거 평점 이력을 바탕으로 리뷰가 조작되었는지 또는 리뷰가 도움이 되는지 판단하는 경향이 있음을 보여준다. 본 연구는 귀인이론을 바탕으로 리뷰어의 과거 평점 이력이 리뷰 조작성에 대한 인식과 리뷰 유용성에 미치는 영향을 분석하여, 해당 연구분야에 새로운 관점을 추가한 기여점이 있다.

• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.347-364
• /
• 2022

Fraudulent companies or sellers strategically manipulate reviews to influence customers' purchase decisions; therefore, the reliability of reviews has become crucial for customer decision-making. Since customers increasingly rely on online reviews to search for more detailed information about products or services before purchasing, many researchers focus on detecting manipulated reviews. However, the main problem in detecting manipulated reviews is the difficulties with obtaining data with manipulated reviews to utilize machine learning techniques with sufficient data. Also, the number of manipulated reviews is insufficient compared with the number of non-manipulated reviews, so the class imbalance problem occurs. The class with fewer examples is under-represented and can hamper a model's accuracy, so machine learning methods suffer from the class imbalance problem and solving the class imbalance problem is important to build an accurate model for detecting manipulated reviews. Thus, we propose an OpenAI-based reviews generation model to solve the manipulated reviews imbalance problem, thereby enhancing the accuracy of manipulated reviews detection. In this research, we applied the novel autoregressive language model - GPT-3 to generate reviews based on manipulated reviews. Moreover, we found that applying GPT-3 model for oversampling manipulated reviews can recover a satisfactory portion of performance losses and shows better performance in classification (logit, decision tree, neural networks) than traditional oversampling models such as random oversampling and SMOTE.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.9 no.5
• /
• pp.1234-1241
• /
• 2008

Cost effectiveness is greatly related with the degree of reducing the testing cost by the technical reviews. In this paper, we present a new metric My for evaluating the cost effectiveness of technical reviews during software development. First, we estimate and compare My with conventional measure using data collected during practical software development procedure, then we show the validity and usefulness of the proposed measure My. Also by formulating the relationship between the data collected during the reviews and the test, we present a method to estimate the value of the metric My using only the data collected during review phase.

• The Journal of the Korea Contents Association
• /
• v.18 no.5
• /
• pp.367-375
• /
• 2018

The study aims to determine which groups leave longer(more active) online reviews(comments) on the film by separating groups, one that satisfied with the movie while the other group dissatisfied with the movie. The data used were rating scores and reviews(comments) from Naver Movie API, and break-even point data provided by Korea Film Commission. We analyzed the relationship between movie rating and review length, before and after movie opening, the characteristics of review length according to the box office, and whether the movie rating affects the review length.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.8
• /
• pp.39-46
• /
• 2021

When conducting social big data analysis for franchise stores, reviews of multiple branches of a franchise can be collected together, from which analysis results can be distorted significantly. To improve its accuracy, it should be possible to filter reviews of other branches properly which are not subject to the analysis. This paper presents a method for social big data analysis which reflects characteristics of franchise stores. The proposed method consists of search key configuration and review filtering. For the former, the open data provided by Small Business Promotion Agency is used to extract region names for collecting reviews more accurately. For the latter, open search APIs provided by Naver or Kakao are used to obtain franchise branch information for filtering reviews of other branches that are not subject to analysis. To verify performance of the proposed method, experiments were conducted based on real social reviews collected from online, where the results showed that the accuracy of the proposed review filtering was 93.6% on the average.

• KIISE Transactions on Computing Practices
• /
• v.22 no.7
• /
• pp.338-343
• /
• 2016

Today, opinion reviews on the Web are often used as a means of information exchange. As the importance of opinion reviews continues to grow, the number of issues for opinion spam also increases. Even though many research studies on detecting spam reviews have been conducted, some limitations of gold-standard datasets hinder research. Therefore, we introduce a new dataset called "Paraphrased Opinion Spam (POS)" that contains a new type of review spam that imitates truthful reviews. We have noticed that spammers refer to existing truthful reviews to fabricate spam reviews. To create such a seemingly truthful review spam dataset, we asked task participants to paraphrase truthful reviews to create a new deceptive review. The experiment results show that classifying our POS dataset is more difficult than classifying the existing spam datasets since the reviews in our dataset more linguistically look like truthful reviews. Also, training volume has been found to be an important factor for classification model performance.

• Journal of the Korea Convergence Society
• /
• v.8 no.4
• /
• pp.1-8
• /
• 2017

In the modern society, the size of the fashion market is continuously increasing both overseas and domestic. When purchasing a product through e-commerce, the evaluation data for the product created by other consumers has an effect on the consumer's decision to purchase the product. By analysing the consumer's evaluation data on the product the company can reflect consumer's opinion which can leads to positive affect of performance to company. In this paper, we propose a method to construct a model to analyze user's sentiment using word embedding space formed by learning review data of amazon fashion products. Experiments were conducted by learning three SVM classifiers according to the number of positive and negative review data using the formed word embedding space which is formed by learning 5.7 million Amazon review data.. Experimental results showed the highest accuracy of 88.0% when learning SVM classifier using 50,000 positive review data and 50,000 negative review data.

• The Journal of Society for e-Business Studies
• /
• v.21 no.2
• /
• pp.151-161
• /
• 2016

In the era of the Web 2.0, characterized by the openness, sharing and participation, it is easy for internet users to produce and share the data. The amount of the unstructured data which occupies most of the digital world's data has increased exponentially. One of the kinds of the unstructured data called personal online product reviews is necessary for both the company that produces those products and the potential customers who are interested in those products. In order to extract useful information from lots of scattered review data, the process of collecting data, storing, preprocessing, analyzing, and drawing a conclusion is needed. Therefore we introduce the text-mining methodology for applying the natural language process technology to the text format data like product review in order to carry out extracting structured data by using R programming. Also, we introduce the data-mining to derive the purpose-specific customized information from the structured review information drawn by the text-mining.

• The Journal of Society for e-Business Studies
• /
• v.26 no.4
• /
• pp.95-118
• /
• 2021

As the AI speaker business has risen significantly in recent years, the potential for numerous uses of AI speakers has gotten a lot of attention. Consumers have created an environment in which they can express and share their experiences with products through various channels, resulting in a large number of reviews that leave consumers with a variety of candid opinions about their experiences, which can be said to be very useful in analyzing consumers' thoughts. Using this review data, this study aimed to examine the factors driving the continued use of AI speakers. Above all, it was determined whether the seven characteristics associated with the intention to adopt AI identified in prior studies appear in consumer reviews. Based on customer review data on Amazon.com, text mining and social network analysis were utilized to examine Amazon eco-products. CONCOR analysis was used to classify words with similar connectivity locations, and Connection centrality analysis was used to classify the factors influencing the continuous use of AI speakers, focusing on the connectivity between words derived by classifying review data into positive and negative reviews. Consumers regarded personality and closeness as the most essential characteristics impacting the continued usage of AI speakers as a result of the favorable review survey. These two parameters had a strong correlation with other variables, and connectedness, in addition to the components established from prior studies, was a significant factor. Furthermore, additional negative review research revealed that recognition failures and compatibility are important problems that deter consumers from utilizing AI speakers. This study will give specific solutions for consumers to continue to utilize Amazon eco products based on the findings of the research.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.13-16
• /
• 2024

온라인상에 인간의 감성을 담은 리뷰 데이터가 꾸준히 축적되어왔다. 이 텍스트 데이터를 분석하고 활용하는 일은 마케팅에 있어서 중요한 자산이 될 것이다. 이와 관련된 Aspect-Based Sentiment Analysis(ABSA) 연구는 한글에 있어서는 데이터 부족을 이유로 거의 선행연구가 없는 실정이다. 본 연구에서는 최근 공개된 데이터 셋을 바탕으로 하여 화장품 도메인에 대한 소비자들의 리뷰 텍스트와 사전 라벨링 된 속성, 감성 극성을 기반으로 ABSA를 진행한다. Klue RoBERTa base 모델을 활용하여 데이터를 학습시키고, Python Kiwipiepy 등으로 전처리한 결과를 대시보드로 시각화하여 분석하기 쉬운 환경을 마련하는 방법을 제시한다.