• 제목/요약/키워드: 상품구매 성향

검색결과 76건 처리시간 0.022초

Resolving the 'Gray sheep' Problem Using Social Network Analysis (SNA) in Collaborative Filtering (CF) Recommender Systems (소셜 네트워크 분석 기법을 활용한 협업필터링의 특이취향 사용자(Gray Sheep) 문제 해결)

  • Kim, Minsung;Im, Il
    • Journal of Intelligence and Information Systems
    • /
    • 제20권2호
    • /
    • pp.137-148
    • /
    • 2014
  • Recommender system has become one of the most important technologies in e-commerce in these days. The ultimate reason to shop online, for many consumers, is to reduce the efforts for information search and purchase. Recommender system is a key technology to serve these needs. Many of the past studies about recommender systems have been devoted to developing and improving recommendation algorithms and collaborative filtering (CF) is known to be the most successful one. Despite its success, however, CF has several shortcomings such as cold-start, sparsity, gray sheep problems. In order to be able to generate recommendations, ordinary CF algorithms require evaluations or preference information directly from users. For new users who do not have any evaluations or preference information, therefore, CF cannot come up with recommendations (Cold-star problem). As the numbers of products and customers increase, the scale of the data increases exponentially and most of the data cells are empty. This sparse dataset makes computation for recommendation extremely hard (Sparsity problem). Since CF is based on the assumption that there are groups of users sharing common preferences or tastes, CF becomes inaccurate if there are many users with rare and unique tastes (Gray sheep problem). This study proposes a new algorithm that utilizes Social Network Analysis (SNA) techniques to resolve the gray sheep problem. We utilize 'degree centrality' in SNA to identify users with unique preferences (gray sheep). Degree centrality in SNA refers to the number of direct links to and from a node. In a network of users who are connected through common preferences or tastes, those with unique tastes have fewer links to other users (nodes) and they are isolated from other users. Therefore, gray sheep can be identified by calculating degree centrality of each node. We divide the dataset into two, gray sheep and others, based on the degree centrality of the users. Then, different similarity measures and recommendation methods are applied to these two datasets. More detail algorithm is as follows: Step 1: Convert the initial data which is a two-mode network (user to item) into an one-mode network (user to user). Step 2: Calculate degree centrality of each node and separate those nodes having degree centrality values lower than the pre-set threshold. The threshold value is determined by simulations such that the accuracy of CF for the remaining dataset is maximized. Step 3: Ordinary CF algorithm is applied to the remaining dataset. Step 4: Since the separated dataset consist of users with unique tastes, an ordinary CF algorithm cannot generate recommendations for them. A 'popular item' method is used to generate recommendations for these users. The F measures of the two datasets are weighted by the numbers of nodes and summed to be used as the final performance metric. In order to test performance improvement by this new algorithm, an empirical study was conducted using a publically available dataset - the MovieLens data by GroupLens research team. We used 100,000 evaluations by 943 users on 1,682 movies. The proposed algorithm was compared with an ordinary CF algorithm utilizing 'Best-N-neighbors' and 'Cosine' similarity method. The empirical results show that F measure was improved about 11% on average when the proposed algorithm was used

    . Past studies to improve CF performance typically used additional information other than users' evaluations such as demographic data. Some studies applied SNA techniques as a new similarity metric. This study is novel in that it used SNA to separate dataset. This study shows that performance of CF can be improved, without any additional information, when SNA techniques are used as proposed. This study has several theoretical and practical implications. This study empirically shows that the characteristics of dataset can affect the performance of CF recommender systems. This helps researchers understand factors affecting performance of CF. This study also opens a door for future studies in the area of applying SNA to CF to analyze characteristics of dataset. In practice, this study provides guidelines to improve performance of CF recommender systems with a simple modification.

  • The effect of the decision to use innovative services on the choice of consumers with a risk-averse tendency (혁신 서비스 이용 결정이 위험회피 성향 소비자의 선택에 미치는 영향)

    • Park, Kikyoung
      • Journal of Service Research and Studies
      • /
      • 제13권2호
      • /
      • pp.146-160
      • /
      • 2023
    • The spread of non-face-to-face services due to the COVID-19 pandemic has brought many changes in consumers' purchasing behavior and attracted much attention to new services. Could trying new services caused by this sudden environmental change alter consumers's choice patterns? This study proposes the research question of whether these new service experiences can change consumers' existing choice behavior, especially for risk-averse consumers who maintain their existing choice behavior or prefer safe alternatives. In this study, we examined whether trying out an unmanned payment services, one of innovative services that emerged after the pandemic crisis, can change the existing choice behavior of risk-averse consumers, i.e., make them more likely to prefer risky alternatives to safe alternatives. To accomplish these research goals, this research conducted one pilot survey and one study. The results of pilot survey showed that the stronger the prevention-focus tendency, the lower the self-efficacy to use the innovative service, with a negative relationship between them. Based on these findings, the study used an experimental method to examine the interaction effects between the use of innovation services and consumers' regulatory focus in a choice behavior and to explore the psychological mechanisms behind them. According to the results, it is found that prevention-focused consumers were more likely to choose risky alternatives and dissimilar extended brands following a trial of an unmanned payment service compared to not using that service. In contrast, promotion-focused consumers did not show different choice patterns regardless of following a trial of an innovative service. Furthermore, these results for prevention-focused consumers confirm the role of self-efficacy as a psychological mechanism. These findings shed light on the role of self-efficacy which has discussed in positive psychology into marketing area. Moreover, practical and academic implications are suggested by the finding that behavioral change occurs in risk-averse consumers, who are known to be hesitant to try new behaviors, indicating market expansion related to potential consumers for the use of the innovation services.

    Chinese Consumer Preference of Chicken Burgers Cooked by Sous-vide with Korean-styled Seasoning and Available on the Chinese Fast Food Market (진공저온조리를 이용한 한식 스타일 치킨버거와 중국 시판 치킨버거의 중국 소비자 기호도)

    • Bae, Sungeun;Jang, Jin A;Oh, Jieun;Lee, Kyungwon;Cho, Mi Sook
      • Korean Journal of Food Science and Technology
      • /
      • 제45권1호
      • /
      • pp.126-132
      • /
      • 2013
    • The purpose of this study is to examine the difference of Chinese consumer preference between chicken burgers cooked by Sous-vide adding Korean-styled seasoning and chicken burgers from local fast food restaurants. For Chinese women in their 20s who reside in Beijing, China, the most important attribute for eating-out was taste (M=6.5), health (M=6.0) and pleasure (M=5.6) in order and 68.60% of them replied that they were interested in Korean cuisine. More than 90% answered they like Korean food for its good taste, shape and color. As for burger preference, BB was significantly higher in terms of the overall and appearance preference. However, there was no significant difference between AB and SB. As for flavor and texture, it showed no significant differences among BB, AB, and SB products. For SB, total balance, soft texture of patty and balance of flavor characteristics were the reasons for their preference of the burger.

    Study on the Relationships Among Perceived Shopping Values, Brand Equity, and Store Loyalty of Korean and Chinese Consumers: A Case of Large Discount Store (한국과 중국 소비자의 쇼핑 경험가치 지각과 브랜드자산 및 점포충성도의 관계에 관한 비교 연구: 대형 할인점을 중심으로)

    • Hwang, Soonho;Oh, Jongchul;Yoon, Sungjoon
      • Asia Marketing Journal
      • /
      • 제14권2호
      • /
      • pp.209-237
      • /
      • 2012
    • 1. Research Purpose Consumers rely on various clues to evaluate their decision to patronize a retail store, and store brand is one of them (Dodds 1991; Grewal et al. 1998). As consumers find ever increasing variety of contact points connecting them to specific store, the value of experiential shopping as a means of increasing store's brand equity warrants greater attention from scholars of retail management. Retail shopping values are credited for creating not only cognitive experiences like brand knowledge but also emotional experiences such as shopping pleasure and pride (Schmitt 1999). This may be because today's consumers place emphasis on emotional values associated with shopping pleasure, lifestyle brought to life, brand relationship, and store atmosphere more than utilitarian values such as product quality and price. Many previous literature found this to be true (Ahn and Lee 2011; Mathwick et al. 2001). This brings forth important research issues and questions regarding the roles of shopping experiential values and brand equity with regard to consumer's retail patronage choice. However, despite this importance, research on this area remains quite inadequate (Hwang 2010). For this reason, this study aims to verify the relationships among experiential shopping values, retail store brand equity and tries to link that with customer loyalty by surveying large-scale discount store shoppers in Korea and China. 2. Research Contents In order to carry out the research objective, this study conducted comprehensive literature survey on previous literature by discussing major findings and implications with regard to shopping values and retail brand equity and store loyalty. For data collection, researcher employed survey-based research method where data were collected in two major cities of Korea (Seoul) and China (Bejing) and sampling frame was based on patrons of large discount stores in both countries. Specific research questions raised in this study are as follows; RQ1: How do Korean and Chinese consumers differently perceive of shopping values regarding shopping at large-sclae discount stores? RQ2: Are there differences in consumers' emotional consumption propensities? RQ3: Do Korean and Chinese consumers display different perceptions of brand equity towards large-scale discount stores? RQ4: Are there differences in relationships between shopping values and brand equity for Korean and Chinese consumers? For statistical analysis, SPSS17.0, AMOS17.0 and SmartPLS were employed. 3. Research Results The data collected through face-to-face survey conducted in Seoul and Bejing revealed appropriate data validity and reliability as a result of exploratory/confirmatory factor analysis and reliability tests, andh SEM model yielding satisfactory model fitness. The result of the study may be summarized by three main points. First, as a result of testing differences in consumption dispositions, Chinese consumers showed higher scores in aesthetic and symbolic dispositions, whereas Korean consumers scored higher in hedonic disposition. Second, testing on perceptions toward brand equity of large discount stores showed that Korean consumers exhibited more positive perceptions of brand awareness and brand image than Chinese counterparts. Third, the result of exploratory factor analysis on the experiential shopping values revealed different factors for each country. On Korean side, consumer interest value, aesthetic value, and hedonic value were prominent, whereas on Chinese side, hedonic value, aesthetic value, consumer interest value, and service excellence value were found salient. 4. Research Implications While many previous studies on inter-country differences in retailing area mainly focused on cultural dispositions or orientations to explain the differences, this study sets itself apart by specifically targeting individual consumer's shopping values from an experiential viewpoint. The study result provides important theoretical as well as practical implications for large-scale discount store, especially the impotance of fully exploring the linkage between shopping values and brand equity, which has significant influence on loyalty. Therefore, the specific implications deriving from the result shed some important insights upon the consumption values based on shopping experiences and brand equity. The differences found in store shoppers between the two countries may also provide useful insights for Korean and Chinese retailers who plan to expand their operations globally. Related strategic implications derived from this study is the importance of localizing retail strategy which is based on the differences found in experiential shopping values between the two country groups. Especially the finding that Chinese consumers value consumer interest and service excellence, whereas Koreans place importance on hedonic or aesthetic values indicates the need to differentiate the consumer's psychographical profiles when it comes to expanding retail operations globally. Particularly important will be to pursue price-orienated strategy in China in consideration of the high emphasis on consumer interests and service excellence, but to emphasize the symbolic aspects of brand equity in Korea by maximizing the brand equity associated with aesthetic values and hedonic orientations. 5. Recommendations This study focused on generic retail branded discount stores in both countries, thus making it difficult to tease out store-specific strategies based on specific retail brands. Future studies may benefit fro employing actual brand names in survey questionnaire to verify relationship between shopping values and brand-based store strategy. As with other studies of this nature, this study needs to strengthen the result's generalizability by selecting respondents from a wider spectrum of respondents.

    • PDF

    A Study on E-mail Campaigns and Feedback Analysis as Marketing Tools of Internet Fashion Shopping Malls - With Focus on Specialized Fashion Shopping Malls - (인터넷 패션쇼핑몰의 이메일 마케팅 활용과 반응 - 패션 전문몰을 중심으로 -)

    • Han, Ji-Sook
      • Archives of design research
      • /
      • 제19권2호
      • /
      • pp.53-62
      • /
      • 2006
    • E-mail has indeed developed from 'a means of instant communication' to an indispensable part of online marketing. Therefore, companies need to implement consistent customer management. Communication with customers and marketing through e-mail is a powerful way of communication and adapting one-to-one marketing strategies to customer trends, habits and taste preferences. Since setting accurate targets is especially important in the fashion industry, e-mail marketing is the most effective way to communicate with customers and one-to-one marketing constitutes a very important strategy. In this study, I will analyze this powerful one-on-one marketing tool, particularly actual e-mail messages sent by an Internet Shopping Mall from June 12 to July 30, 2005, examine the effect of these messages on sales growth and analyze actual feedback received. Regarding e-mail read rates broken down by age and gender, 1 found that females in their late twenties recorded the highest rate at 21.66% and their contribution to sales growth was recorded at 3.5% From actual sales records, found that 28.10% of total sales were attributable to people in their late twenties, showing that the age group that reads e-mails the most also buys the most. Regarding feedback by e-mail title, e-mails from the 'Casual' category seemed to be the most effective, in that most of these e-mails were read. Also, messages sent on Tuesdays were read the most, according to the feedback analysis by weekday. Section e-mails were read more often than regular e-mails. Regarding the view rate according to the time e-mails were sent, messages sent to females in their late twenties at two o'clock in the afternoon were read by 20.93% of recipients, recording the highest read rate. By offering informative content and practical tips, visitors will be attracted to the site and generate site traffic. Therefore, we can conclude that sending e-mail messages can greatly contribute to sales growth and e-mail marketing is very effective. Also, in order to make e-mail campaigns more effective and improve marketing results, we need to analyze actual results and apply our findings in future e-mail campaigns. With this, we get successful marketing results.

    • PDF

    Incorporating Social Relationship discovered from User's Behavior into Collaborative Filtering (사용자 행동 기반의 사회적 관계를 결합한 사용자 협업적 여과 방법)

    • Thay, Setha;Ha, Inay;Jo, Geun-Sik
      • Journal of Intelligence and Information Systems
      • /
      • 제19권2호
      • /
      • pp.1-20
      • /
      • 2013
    • Nowadays, social network is a huge communication platform for providing people to connect with one another and to bring users together to share common interests, experiences, and their daily activities. Users spend hours per day in maintaining personal information and interacting with other people via posting, commenting, messaging, games, social events, and applications. Due to the growth of user's distributed information in social network, there is a great potential to utilize the social data to enhance the quality of recommender system. There are some researches focusing on social network analysis that investigate how social network can be used in recommendation domain. Among these researches, we are interested in taking advantages of the interaction between a user and others in social network that can be determined and known as social relationship. Furthermore, mostly user's decisions before purchasing some products depend on suggestion of people who have either the same preferences or closer relationship. For this reason, we believe that user's relationship in social network can provide an effective way to increase the quality in prediction user's interests of recommender system. Therefore, social relationship between users encountered from social network is a common factor to improve the way of predicting user's preferences in the conventional approach. Recommender system is dramatically increasing in popularity and currently being used by many e-commerce sites such as Amazon.com, Last.fm, eBay.com, etc. Collaborative filtering (CF) method is one of the essential and powerful techniques in recommender system for suggesting the appropriate items to user by learning user's preferences. CF method focuses on user data and generates automatic prediction about user's interests by gathering information from users who share similar background and preferences. Specifically, the intension of CF method is to find users who have similar preferences and to suggest target user items that were mostly preferred by those nearest neighbor users. There are two basic units that need to be considered by CF method, the user and the item. Each user needs to provide his rating value on items i.e. movies, products, books, etc to indicate their interests on those items. In addition, CF uses the user-rating matrix to find a group of users who have similar rating with target user. Then, it predicts unknown rating value for items that target user has not rated. Currently, CF has been successfully implemented in both information filtering and e-commerce applications. However, it remains some important challenges such as cold start, data sparsity, and scalability reflected on quality and accuracy of prediction. In order to overcome these challenges, many researchers have proposed various kinds of CF method such as hybrid CF, trust-based CF, social network-based CF, etc. In the purpose of improving the recommendation performance and prediction accuracy of standard CF, in this paper we propose a method which integrates traditional CF technique with social relationship between users discovered from user's behavior in social network i.e. Facebook. We identify user's relationship from behavior of user such as posts and comments interacted with friends in Facebook. We believe that social relationship implicitly inferred from user's behavior can be likely applied to compensate the limitation of conventional approach. Therefore, we extract posts and comments of each user by using Facebook Graph API and calculate feature score among each term to obtain feature vector for computing similarity of user. Then, we combine the result with similarity value computed using traditional CF technique. Finally, our system provides a list of recommended items according to neighbor users who have the biggest total similarity value to the target user. In order to verify and evaluate our proposed method we have performed an experiment on data collected from our Movies Rating System. Prediction accuracy evaluation is conducted to demonstrate how much our algorithm gives the correctness of recommendation to user in terms of MAE. Then, the evaluation of performance is made to show the effectiveness of our method in terms of precision, recall, and F1-measure. Evaluation on coverage is also included in our experiment to see the ability of generating recommendation. The experimental results show that our proposed method outperform and more accurate in suggesting items to users with better performance. The effectiveness of user's behavior in social network particularly shows the significant improvement by up to 6% on recommendation accuracy. Moreover, experiment of recommendation performance shows that incorporating social relationship observed from user's behavior into CF is beneficial and useful to generate recommendation with 7% improvement of performance compared with benchmark methods. Finally, we confirm that interaction between users in social network is able to enhance the accuracy and give better recommendation in conventional approach.


    (34141) Korea Institute of Science and Technology Information, 245, Daehak-ro, Yuseong-gu, Daejeon
    Copyright (C) KISTI. All Rights Reserved.