您的位置：首页 > 论文页面
教学视频精准推荐算法研究
发表时间：2017-11-30 浏览量：2672 下载量：484
| 全部作者： | 尹杭，宋文广，龙力，吴华超 |
| 作者单位： | 长江大学计算机科学学院 |
| 摘 要： | 通过分析现有精准营销推荐算法，结合学员通过网络学习教学视频的特点，研究传统协同过滤算法的优缺点，提出一种基于教学视频的协同过滤算法。首先设计一种基于教学视频网站的聚类协同过滤算法，使用随机函数在样本中初始化多个聚类中心。再将样本分配给距离其最近的中心向量，由这些样本构造不相交的聚类，用各个聚类的中心向量作为新的中心。最后进行循环计算分组和确定中心的步骤，直至算法收敛或到达确定的迭代步数为止。通过分析实验数据得出，经过聚类后的样本数据在推荐过程中表现出更高的价值。该算法能提高学员搜索相关教学视频的速度，并能提高学习效率。 |
| 关 键 词： | 算法理论；教学视频；协同过滤；精准推荐 |
| Title： | Research on accurate recommendation algorithm for teaching video |
| Author： | YIN Hang, SONG Wenguang, LONG Li, WU Huachao |
| Organization： | School of Computer Science, Yangtze University |
| Abstract： | By analyzing the existing accurate marketing recommendation algorithm, combined with the characteristics of students learning online teaching video, this paper studies the advantages and disadvantages of traditional collaborative filtering algorithm, and proposes a collaborative filtering algorithm based on teaching video. Firstly, a clustering collaborative filtering algorithm based on the teaching video website is designed, which uses the random function to initialize multiple cluster centers in the sample. Then, the samples are assigned to the nearest central vectors, the disjoint clusters are constructed by these samples, and the central vectors of each cluster are taken as the new centers. Finally, the steps of grouping and determining the center are calculated circularly until the algorithm converges or arrives at the determined iteration step. By analyzing the experimental data, the sample data after clustering show higher value in the recommendation process. The algorithm can improve the speed of searching for related teaching video by students, and also improve the learning efficiency. |
| Key words： | algorithm theory; teaching video; collaborative filtering; accurate recommendation |
| 发表期数： | 2017年11月第22期 |
| 引用格式： | 尹杭，宋文广，龙力，等. 教学视频精准推荐算法研究[J]. 中国科技论文在线精品论文，2017，10（22）：2445-2451. |
0
评论数 0请您登录
暂无评论
下载全文
多维论文