Jin Xu

This study was conducted during the development of a spatial augmented reality (SAR) curling game, with the aim of selecting an appropriate interaction method. Currently, pointer-based interaction, gesture interaction, controller interaction, and tangible interaction are the main interaction methods used in sports games. Sports games emphasize the realism of movement and the speed of response, which are greatly impacted by the naturalness, performance, and usability of the interaction method. However, their differences concerning the aforementioned metrics and impacts on the gaming experience remain unclear. Therefore, a comparative study was conducted to evaluate the interaction performance, usability, and gaming experience offered by these four methods. Specifically, an SAR prototype system supporting selection, alignment, and system control tasks was developed with a user study conducted involving 40 undergraduate or graduate students. The results showed that in terms of interaction performance, tangible interaction was the most efficient and with high accuracy, slightly lower than the most familiar mouse method. In terms of usability, tangible interaction scored the highest. In terms of gaming experience, tangible interaction received the highest overall evaluation. Consequently, tangible interaction was applied to the curling game, with the hope of extending its use to more sports games. Hopefully, this study can provide valuable insights and guidance for other sports games in the selection of interaction methods.

As the cornerstone of spatial geometry and engineering drawings, three‐view drawing lays the necessary foundation for students to learn geography, graphics, industrial design, geographic information science, and so forth. However, teaching three‐view drawing can be challenging due to its abstraction. Currently, desktop and handheld augmented reality (AR) systems are widely applied to three‐view teaching. However, existing systems suffer from visuo‐tactile inconsistency and contain unfriendly interactions, resulting in a poor learning experience. In this paper, head‐mounted AR is introduced for the first time in the teaching of three‐view drawing. A system was designed for learning three‐view drawing to facilitate students' free exploration and learning in lessons. Additionally, a comparative experiment was designed and conducted to investigate whether multimodal interaction and head‐mounted AR help students to better understand the relationship between concrete three‐dimensional (3D) objects and abstract two‐dimensional planes and the impact on subjective perceptions. The results demonstrated that the proposed system significantly improved learning performance. Furthermore, the proposed system heightened students' interest in learning, made reproducing 3D structures more intuitive, and students were more willing to use such systems in their future studies.

Jialong Tang, Hongyu Lin, Meng Liao, Yaojie Lu, Xianpei Han, Le Sun, Weijian Xie, Jin Xu. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

In the neighborhood-based collaborative filtering recommendation algorithm, the accuracy of the similarity calculation determines the quality of the recommendation algorithm directly. The traditional similarity measure only considers influence of common rated items among users, and ignores the attribute characteristics of users' rated items. Low-precision similarity metrics reduce performance of recommended systems, when the dataset is extremely sparse. In order to solve above problems, this paper proposes a similarity measure model considering users' preferences for item attributes. The model fully considers the user's preferences for item attributes and co-rated items, and the number of co-rated items. The model establishes more connections between users and items, so as to mine user interests effectively and make it more in line with the actual application. The experimental results show that the model proposed by this paper is superior to other comparison methods in accuracy and diversity, which effectively improves the performance of the recommended algorithm.