A Multi-CNN Model Interacting Contents and Ratings for Predicting Review Helpfulness

Xinzhe Li,Qinglong Li,Jaekyeong Kim 한국지능정보시스템학회 2022 한국지능정보시스템학회 학술대회논문집 Vol.2022 No.6

With the growth of the e-commerce industry, online consumer reviews significantly impact the consumer purchase decision process. Since the consistently increasing number of reviews, the consumer can face an information overload problem. Thus, the consumers have a challenge exploring the information they need. Thus, we argue that predicting the review helpfulness becomes significant. When predicting review helpfulness, since the review contents and star ratings are information written from the same consumer experience, the consistency of the review contents and star ratings is essential. Previous studies predict review helpfulness by considering review content and star ratings simultaneously. However, such an approach has limitations in the representation capacity of star ratings and the capture of the interaction between review content and star ratings. The current study proposed a CNN-CRI mechanism to address the limitations of the previous study. To evaluate the proposed methodology, we utilized real-world online review data from Amazon.com. The results show that our study model indicates better performance than the state-of-the-art approach.

명시적 및 암시적 피드백을 활용한 그래프 컨볼루션 네트워크 기반 추천 시스템 개발

이흠철(Xinzhe Li),김동언(Dongeon Kim),이청용(Qinglong Li),김재경(JaeKyeong Kim) 한국IT서비스학회 2023 한국IT서비스학회지 Vol.22 No.1

With the development of the e-commerce market, various types of products continue to be released. However, customers face an information overload problem in purchasing decision-making. Therefore, personalized recommendations have become an essential service in providing personalized products to customers. Recently, many studies on GCN-based recommender systems have been actively conducted. Such a methodology can address the limitation in disabling to effectively reflect the interaction between customer and product in the embedding process. However, previous studies mainly use implicit feedback data to conduct experiments. Although implicit feedback data improves the data scarcity problem, it cannot represent customers' preferences for specific products. Therefore, this study proposed a novel model combining explicit and implicit feedback to address such a limitation. This study treats the average ratings of customers and products as the features of customers and products and converts them into a high-dimensional feature vector. Then, this study combines ID embedding vectors and feature vectors in the embedding layer to learn the customer-product interaction effectively. To evaluate recommendation performance, this study used the MovieLens dataset to conduct various experiments. Experimental results showed the proposed model outperforms the state-of-the-art. Therefore, the proposed model in this study can provide an enhanced recommendation service for customers to address the information overload problem.

CNN 기반 리뷰 유용성 점수 예측을 통한 개인화 추천 서비스 성능 향상에 관한 연구

이청용(Qinglong Li),이병현(Byunghyun Lee),이흠철(Xinzhe Li),김재경(Jae Kyeong Kim) 한국지능정보시스템학회 2021 지능정보연구 Vol.27 No.3

Recently, various types of products have been launched with the rapid growth of the e-commerce market. As a result, many users face information overload problems, which is time-consuming in the purchasing decision-making process. Therefore, the importance of a personalized recommendation service that can provide customized products and services to users is emerging. For example, global companies such as Netflix, Amazon, and Google have introduced personalized recommendation services to support users purchasing decisions. Accordingly, the users information search cost can reduce which can positively affect the companys sales increase. The existing personalized recommendation service research applied Collaborative Filtering (CF) technique predicts user preference mainly use quantified information. However, the recommendation performance may have decreased if only use quantitative information. To improve the problems of such existing studies, many studies using reviews to enhance recommendation performance. However, reviews contain factors that hinder purchasing decisions, such as advertising content, false comments, meaningless or irrelevant content. When providing recommendation service uses a review that includes these factors can lead to decrease recommendation performance. Therefore, we proposed a novel recommendation methodology through CNN-based review usefulness score prediction to improve these problems. The results show that the proposed methodology has better prediction performance than the recommendation method considering all existing preference ratings. In addition, the results suggest that can enhance the performance of traditional CF when the information on review usefulness reflects in the personalized recommendation service.

Seismic analysis of half-through steel truss arch bridge considering superstructure

Ruiqi Li,Xinzhe Yuan,Wancheng Yuan,Xinzhi Dang,Guoyu Shen 국제구조공학회 2016 Structural Engineering and Mechanics, An Int'l Jou Vol.59 No.3

This paper takes a half-through steel truss arch bridge as an example. A seismic analysis is conducted with nonlinear finite element method. Contrast models are established to discuss the effect of simplified method for main girder on the accuracy of the result. The influence of seismic wave direction and wave-passage on seismic behaviors are analysed as well as the superstructure and arch ring interaction which is mostly related with the supported bearings and wind resistant springs. In the end, the application of cable-sliding aseismic devices is discussed to put forward a layout principle. The main conclusions include: ① The seismic response isn’t too distinctive with the simplified method of main girder. Generally speaking, the grillage method is recommended. ② Under seismic input from different directions, arch foot is usually the mostly dangerous section. ③ Vertical wave input and horizontal wave-passage greatly influence the seismic responses of arch ring, significantly increasing that of midspan. ④ The superstructure interaction has an obvious impact on the seismic performance. Half-through arch bridges with long spandrel columns fixed has a less response than those with short ones fixed. And a large stiffness of wind resistant spring makes the the seismic responses of arch ring larger. ⑤ A good isolation effectiveness for half-through arch bridge can be achieved by a reasonable arrangement of CSFABs.

The Biocompatibility of Multi-Source Stem Cells and Gelatin-Carboxymethyl Chitosan-Sodium Alginate Hybrid Biomaterials

Wang Xinzhe,Li Siqi,Yu Honglian,Lv Jianzhi,Fan Minglun,Wang Ximing,Wang Xin,Liang Yanting,Mao Lingna,Zhao Zhankui 한국조직공학과 재생의학회 2022 조직공학과 재생의학 Vol.19 No.3

BACKGROUND: Nowadays, biological tissue engineering is a growing field of research. Biocompatibility is a key indicator for measuring tissue engineering biomaterials, which is of great significance for the replacement and repair of damaged tissues. METHODS: In this study, using gelatin, carboxymethyl chitosan, and sodium alginate, a tissue engineering material scaffold that can carry cells was successfully prepared. The material was characterized by Fourier transforms infrared spectroscopy. In addition, the prepared scaffolds have physicochemical properties, such as swelling ratio, biodegradability. we observed the biocompatibility of the hydrogel to different adult stem cells (BMSCs and ADSCs) in vivo and in vitro. Adult stem cells were planted on gelatin-carboxymethyl chitosan-sodium alginate (Gel/SA/CMCS) hydrogels for 7 days in vitro, and the survival of stem cells in vitro was observed by live/died staining. Gel/SA/CMCS hydrogels loaded with stem cells were subcutaneously transplanted into nude mice for 14 days of in vivo culture observation. The survival of adult stem cells was observed by staining for stem cell surface markers (CD29, CD90) and Ki67. RESULTS: The scaffolds had a microporous structure with an appropriate pore size (about 80 lm). Live/died staining showed that adult stem cells could stably survive in Gel/SA/CMCS hydrogels for at least 7 days. After 14 days of culture in nude mice, Ki67 staining showed that the stem cells supported by Gel/SA/CMCS hydrogel still had high proliferation activity. CONCLUSION: Gel/SA/CMCSs hydrogel has a stable interpenetrating porous structure, suitable swelling performance and degradation rate, can promote and support the survival of adult stem cells in vivo and in vitro, and has good biocompatibility. Therefore, Gel/SA/CMCS hydrogel is a strong candidate for biological tissue engineering materials.

Dynamic numerical analysis of single-support modular bridge expansion joints

Wancheng Yuan,Xinzhe Yuan,Ruiqi Li,Jian’guo Wang 국제구조공학회 2016 Steel and Composite Structures, An International J Vol.22 No.1

Severe fatigue and noise problems of modular bridge expansion joints (MBEJs) are often induced by vehicle loads. However, the dynamic characteristics of single-support MBEJs have yet to be further investigated. To better understand the vibration mechanism of single-support MBEJs under vehicle loads, a 3D finite element model of single-support MBEJ with five center beams is built. Successive vehicle loads are given out and the vertical dynamic responses of each center beams are analyzed under the successive loads. Dynamic amplification factors (DAFs) are also calculated along with increasing vehicle velocities from 20 km/h to 120 km/h with an interval 20 km/h. The research reveals the vibration mechanism of the single-support MBEJs considering coupled center beam resonance, which shows that dynamic responses of a given center beam will be influenced by the neighboring center beams due to their rebound after the vehicle wheels depart. Maximal DAF 1.5 appears at 120 km/h on the second center beam. The research results can be utilized for reference in the design, operation and maintenance of singlesupport MBEJs.