Institute for Advanced Study, Confucius Academy, Guiyang, Guizhou 550025, China.
Comput Intell Neurosci. 2022 Aug 29;2022:7715851. doi: 10.1155/2022/7715851. eCollection 2022.
Chinese traditional culture is the treasure of our cultural field. In the new era, it is of great significance to give traditional culture a new life and vitality. The term "big data" is hotly debated all over the world, while the development of big data is gradually occupying all aspects of the society that people are compatible with society. It is an imperative initiative to build a cultural data system by making use of big data technology, and cultural big data can make Chinese traditional culture release more vitality. This paper analyzes the new characteristics of traditional culture development from big data in helping traditional culture inheritance and innovation and proposes new ideas and creates more possibilities for the development of traditional culture. Combining with big data technology, this paper proposes an improvement to the data sparsity problem and cold-start problem of collaborative filtering recommendation algorithm and also improves the recommendation algorithm based on association rules. The association rule technique is used to compensate for the cold-start and data sparsity problems of new users often encountered by collaborative filtering techniques; the aim is to obtain recommendation results with high user satisfaction. Experiments on traditional cultural resource datasets show that the method in this paper effectively solves the data sparsity and cold-start problems that exist in traditional collaborative filtering techniques, and the recommendation accuracy surpasses that of other methods.
中国传统文化是文化领域的瑰宝。在新时代,赋予传统文化新的生命和活力具有重要意义。“大数据”一词在全球范围内引发热议,而大数据的发展正逐渐占据人们与之兼容的社会的各个方面。利用大数据技术构建文化数据体系,是文化大数据使中国传统文化焕发更多活力的必然之举。本文从大数据视角分析传统文化发展的新特点,为传统文化的传承与创新提供新思路,为传统文化的发展创造更多可能。本文结合大数据技术,对协同过滤推荐算法中的数据稀疏性问题和冷启动问题提出改进,改进基于关联规则的推荐算法。关联规则技术用于弥补协同过滤技术经常遇到的新用户的冷启动和数据稀疏性问题;目的是获得具有高用户满意度的推荐结果。在传统文化资源数据集上的实验表明,本文提出的方法有效地解决了传统协同过滤技术中存在的数据稀疏性和冷启动问题,推荐精度超过其他方法。