• Journal of the Korea Society of Computer and Information
• /
• v.27 no.3
• /
• pp.173-182
• /
• 2022

In this paper, we propose a machine learning model that can prevent fraudulent transactions in advance and interpret them using the XAI approach. For the experiment, we collected a real data set of 12,258 mobile phone sales posts from Joonggonara, a major domestic online C2C resale trading platform. Characteristics of the text corresponding to the post body were extracted using Doc2vec, dimensionality was reduced through PCA, and various derived variables were created based on previous research. To mitigate the data imbalance problem in the preprocessing stage, a complex sampling method that combines oversampling and undersampling was applied. Then, various machine learning models were built to detect fraudulent postings. As a result of the analysis, LightGBM showed the best performance compared to other machine learning models. And as a result of SHAP, if the price is unreasonably low compared to the market price and if there is no indication of the transaction area, there was a high probability that it was a fraudulent post. Also, high price, no safe transaction, the more the courier transaction, and the higher the ratio of 0 in the price also led to fraud.

• The Journal of the Korea Contents Association
• /
• v.22 no.6
• /
• pp.15-23
• /
• 2022

The operation method of offline venues such as local festivals usually has problems such as the absence of real-time events and booth information, wasting time in the process of purchasing and receiving goods, and stagnating the movement of guests. This study increases the convenience of guests and managers by integrating all contents in the venue in the form of booth, thereby increasing the uniformity of event information and the efficiency of booth operation. In addition, a system was designed to minimize movement problems and improve performance by detecting the movement of guests within the venue and increasing the efficiency of arranging booths based on location data. It has been developed as a low-cost system that measures wireless packets with portable wireless LAN APs and control units. This has advantages in the operation of the venue, which consists of the installation and dismantling of variable booths in a short period of time. It is expected that the integrated operation of the venue will be utilized by linking the movement data of guests to activate visits and increase sales through data-based promotions.

• The Journal of the Korea Contents Association
• /
• v.22 no.7
• /
• pp.73-80
• /
• 2022

In this study, prior studies of 280 small business owners in Korea over the past two decades were comprehensively analyzed through keyword network and LDA topic modeling analysis, and overall views and trends in academia were examined. As core keywords, "sales" and "protection," which conflict with each other but are essential for stable and sustainable growth were selected, and 7 topics (Topic 1: start-up, topic 2: digital, topic 3: tax system, topic 4: capability, topic 5: coexistence, topic 6: regulation, and topic 7: funding) were drawn up. Based on the results of the analysis, the need to improve digital maturity for the continued growth and development of small business owners was raised, and the response at the pan-ministerial level and the stability of the performance of functions that can survive even after the new administration to solve the economic damage problems facing small business owners were suggested. In addition, attention to the long-term, speed, detail, and direction of government support in a new way, and a flexible approach to the negative way in which pre-allowance and post-regulation is given were suggested.

• Journal of Digital Convergence
• /
• v.20 no.2
• /
• pp.127-140
• /
• 2022

Recently, accelerators are attracting attention as a new startup incubator. This study empirically analyzed the relative importance and priority of decision-making factors when a startup selects an accelerator using the AHP technique. As a result, startup founders recognized the importance of decision-making factors in the order of funding factors, program factors, and accelerator characteristic factors. there was a difference in the relative importance of detailed factors between startup founders and accelerator workers, depending on the location of the business site and the presence or absence of sales performance. This suggests that accelerators need to strengthen the financial support capabilities and improve customized services to meet the needs of startups. This study can be used as a decision-making criterion when a startup selects an accelerator, and can be used as basic data for additional research on the accelerator program.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.6
• /
• pp.867-871
• /
• 2022

Amid the recent rapid trend change, the change in design has a great impact on the sales of fashion companies, so it is inevitable to be careful in choosing new designs. With the recent development of the artificial intelligence field, various machine learning is being used a lot in the fashion market to increase consumers' preferences. To contribute to increasing reliability in the development of new products by quantifying abstract concepts such as preferences, we generate new images that do not exist through three adversarial generative neural networks (GANs) and numerically compare abstract concepts of preferences using pre-trained convolution neural networks (CNNs). Deep convolutional generative adversarial networks (DCGAN), Progressive growing adversarial networks (PGGAN), and Dual Discriminator generative adversarial networks (DANs), which were trained to produce comparative, high-level, and high-level images. The degree of similarity measured was considered as a preference, and the experimental results showed that D2GAN showed a relatively high similarity compared to DCGAN and PGGAN.

• Fashion & Textile Research Journal
• /
• v.24 no.4
• /
• pp.431-442
• /
• 2022

Despite the rapid reorganization of the center of consumption online, the fashion industry is still strengthening brand marketing using offline stores. This study investigates the psychological mechanisms of shopping flow by three mediators that influence the e-word of mouth(e-WOM) as a marketing performance variable in recent marketing. Data collection was conducted online for 241 women in their 20s and 30s. The significance of multi-mediated pathways was verified using Process 3.5 Model 6. The results for multiple mediation paths are as follows. First, the direct effect of shopping flow on brand advocacy was not significant. Second, analysis of the significance of the indirect effect via simple mediation found that the path mediated by shopping value in the path from the shopping flow to brand advocacy was significant. Third, analysis of the indirect effect by double mediation revealed all three double mediating effects of shopping flow on brand advocacy were significant(emotional response and shopping value, emotional response and store attachment, shopping value and store attachment). Fourth, analysis of the significance of the indirect effect by sequential multiple mediation established the indirect effect leading to emotional response, shopping value, and store attachment was significant. These results indicate that the operation of an experiential fashion store is not just a means for sales, but a communication tool that improves and promotes the brand advocacy by providing brand experience in a store.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.25 no.2
• /
• pp.17-29
• /
• 2023

The digital environment, which has been rapidly developing since the beginning of the 21^st century, has become more specific due to COVID-19, and marketing strategies are rapidly changing to suit purchasing activities of Generation MZ, whose online purchases are becoming the center of their lives. A public relations agency is generally responsible for all aspects of making a client's product or service visible to the public through various forms of media. Among them, a company that performs only fashion-related tasks is called a "fashion PR agency". Now, the fashion industry is also centered on the e-commerce environment, and various digital marketing strategies have been developed and directly related to sales. This study examined the current status of online media and digital marketing, analyzes the aspects of fashion brand promotion strategies and methods in the e-commerce era, focusing on the work of fashion PR agencies, and suggests the direction of new online and offline promotion methods based on marketing and technological aspects. As a result of the study, first, theories on strategies for online media and digital marketing were examined, and found that the amount of online promotion has recently increased and become more specialized. Second, this study examines the concept of fashion PR agencies and analyzed their main tasks through interviews with fashion PR professionals. Third, based on successful online fashion promotion cases, the study analyzed fashion promotion strategies and methods that are being integrated online and offline in the e-commerce era. The main methods included SNS strategy, content strategy, performance strategy, influencer strategy, and event strategy, and it is suggested that integrated management is necessary for consistent brand image management, and an IMC (Integrated Marketing Communication) strategy, which intensively manages all strategies, should be employed.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.323-325
• /
• 2021

The size of the delivery market in Korea is growing year by year. In recent years, the growth rate has skyrocketed due to the aftermath of the coronavirus. Looking at the domestic delivery market's volume trend in 2020, about 3.4 billion boxes increased by 21% compared to about 2.8 billion boxes last year. In addition, sales amounted to 7.5 trillion won, an increase of about 19% compared to 6.3 trillion won a year earlier. As the delivery market grows, the proportion of courier damage relief is also occurring at a considerable rate. About 33% of 1,000 people have experienced delivery accidents, and about 41% of the week have experienced damage or damage. In this paper, a deep learning model capable of detecting a parcel was created to detect a damaged parcel. A system that can check the performance of this model and detect and classify parcels during the delivery process using a real-time detection camera was studied.

• The Journal of Bigdata
• /
• v.6 no.2
• /
• pp.51-60
• /
• 2021

The market size of wearable devices is growing rapidly every year, and manufacturers around the world are introducing products that utilize their unique characteristics to keep up with the demand. Among them, smart watches are wearable devices with a very high share in sales, and they provide a variety of services to users by using information collected in real-time. The quality of service depends on the accuracy of the data collected by the smart watch, but data measurement may not be possible depending on the situation. This paper introduces a method to restore data that a smart watch could not collect. It deals with the similarity calculation method of trajectory information measured over time for data restoration and introduces a procedure for restoring missing sections according to the similarity. To prove the performance of the proposed methodology, a comparative experiment with a machine learning algorithm was conducted. Finally, the expected effects of this study and future research directions are discussed.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.7
• /
• pp.139-148
• /
• 2024

In the post-pandemic era, the demand for online education platforms has surged, leading to increased consumer reliance on online reviews for decision-making. This study investigates the impact of Chinese online customer reviews on consumer purchase behavior in online education. By examining the role of trust, review sentiment, and the quantity and timeliness of reviews, the research aims to understand how these factors influence consumer decisions. By using regression model, findings reveal that negative reviews, timely feedback, and a higher volume of reviews positively affect consumer purchase decisions, while course pricing demonstrates an inverse relationship. Furthermore, cognitive and affective trust mediate the relationship between reviews and purchase behavior, highlighting a reverse U-shaped effect on consumer decision inclination. These insights provide valuable implications for online education providers, emphasizing the need to manage and leverage online reviews to foster consumer trust and improve sales performance.