ISSN 1000-1239 CN 11-1777/TP

    2020 Excellent Young Scholars

    Default Latest Most Read
    Please wait a minute...
    For Selected: Toggle Thumbnails
    Journal of Computer Research and Development    2020, 57 (1): 1-2.  
    Abstract920)   HTML408)    PDF (186KB)(288)       Save
    Related Articles | Metrics
    A Flexible Accuracy-Controllable Searchable Symmetric Encryption Scheme
    Li Ximing, Tao Ruyu, Su Chen, Huang Qiong, Huang Xinyi
    Journal of Computer Research and Development    2020, 57 (1): 3-16.   DOI: 10.7544/issn1000-1239.2020.20190233
    Abstract1090)   HTML30)    PDF (2476KB)(430)       Save
    In the traditional keyword-based searchable symmetric encryption technology, the keyword set is usually generated by the keyword extraction algorithm, so that the content and quantity of the keywords are limited by the keyword extraction algorithm. Therefore, in the keyword-based encryption search system, in addition to the keyword set generated by the system at the time of initial construction, the user cannot search for other related contents, thereby limiting the application of the encryption search technology. In view of the above problems, this paper proposes a flexible accuracy-controllable searchable symmetric encryption (FASSE) that supports flexible and precise control. By flexibly generating keywords and indexes generated by document summary during system operation, the dependence of the keyword collection effectively improves the flexibility of the encryption search technology. FASSE provides three basic searches, namely one-shot search reinforcement search and filter search, which respectively correspond to the user finding the keyword record in the dictionary only once, not finding the keyword record in the dictionary and only using it once. The search finds records or three search cases where keyword records are found in the dictionary and abstract multiple times. At the same time, the system also combines three kinds of search to design a fuzzy reinforcement search to further enhance the practicability of the system. The specific implementation language of the FASSE program is the Java programming development language and the final experiment shows that FASSE averages 114.26 ms in searching each paper in the paper dataset.
    Related Articles | Metrics
    A Survey on Many-Lights Rendering Methods
    Liu Yifan, Xu Kun
    Journal of Computer Research and Development    2020, 57 (1): 17-31.   DOI: 10.7544/issn1000-1239.2020.20190208
    Abstract1045)   HTML78)    PDF (4092KB)(429)       Save
    Many-lights rendering has always been an important research topic in computer graphics. It is one of the important methods to achieve global illumination effects, and it is a great demand in related industries such as games, movies, and animations, etc. However, efficient many-lights rendering is still a big challenge in both fields of off-line rendering and real-time rendering. This paper reviews the main progress of many-lights rendering methods in recent years. Among those many-lights rendering methods, their major goal is to improve the rendering efficiency. In the field of off-line rendering, we first review the algorithms of accelerating visibility test computation for reducing the average rendering time of a single light source. Then, we discuss light source clustering algorithms, and review accelerated rendering algorithms which are directly based on light source clustering. Different strategies for light source clustering are discussed, including strategies using hierarchical structures and strategies based on matrix analysis. After that, we review important sampling methods based on light clustering. In the field of real-time rendering, we introduce several rendering algorithms based on light culling. We also provide comparison and analysis of the advantages and disadvantages of various methods, and summarize the research trends and challenges of many-lights rendering.
    Related Articles | Metrics
    Recent Progress in Large-Scale Ridesharing Algorithms
    Xu Yi, Tong Yongxin, Li Wei
    Journal of Computer Research and Development    2020, 57 (1): 32-52.   DOI: 10.7544/issn1000-1239.2020.20190239
    Abstract1510)   HTML97)    PDF (2479KB)(1243)       Save
    Ridesharing, a new shared mobility service where passengers with different origins and destinations agree to take the same vehicle for their rides and share the cost, is experiencing widespread adoption with the development of sharing economy. In the era of mobile Internet and ubiquitous computing, ridesharing becomes large-scale and exhibits new characteristics including enormous data, dynamic scenarios, diverse objectives, and varied applications. These characteristics fundamentally complicate the ridesharing problem and have spurred extensive new research on large-scale ridesharing algorithms. Furthermore, research on social aspects of large-scale ridesharing has also attracted increasing research attention. This paper aims at a systematic and comprehensive survey on recent advances in large-scale ridesharing algorithms. We first introduce the basic concepts and workflow of ridesharing, and then systematically review existing algorithms for route planning, the core algorithmic problem for large-scale ridesharing. We also discuss social aspects critical for practical large-scale ridesharing applications such as incentive mechanisms, privacy and safeguard measures and finally point out potential future research directions.
    Related Articles | Metrics