Traditional folk culture, which records the track of local historical development, is a historical product reflecting the humanistic style and features and has important historical and cultural value. In the context of big data, analyzing the characteristics of traditional folk culture and excavating the internal relationship and implicit information between traditional folk culture data are the common concerns in the field of traditional cultural information science. Based on the research on the digital characteristics of traditional folk art and the data clustering method under the background of big data, this paper proposes a deep self-organizing map fuzzy C-means (DSOM-FCM) model based on dynamic self-organizing neural network, which integrates resources of traditional folk culture in Shaanxi Province. It not only takes into account the needs of economic development and spiritual civilization construction, but also meets the needs of social and cultural development in Shaanxi Province. Firstly, the spatial and temporal characteristics of big data of folk traditional art are analyzed, and then the input vector of dynamic self-organizing neural network is determined to be 6-dimensional attribute data. Then, based on the traditional self-organizing mapping (SOM) algorithm and fuzzy C-means technology, a traditional folk art resource integration model based on DSOM-FCM is constructed. Finally, using the traditional culture data set test model of Shannxi Province, the experimental results are as follows: When SF = 0.35, the number of clusters of the algorithm is 3, and coarse clustering is realized. When SF = 0.7, the number of clusters of the algorithm is 6, and fine clustering is realized. In order to test the efficiency and accuracy of the model, from the perspectives of classification error, iteration time, number of iterations, and number of clusters, a comparison experiment with SOM and TreeGNG algorithms is set; the results show that the algorithm designed and used in this paper performs well in solving the optimization and integration of traditional folk culture.

中文翻译：

---

基于DSOM-FCM的传统民俗文化优化整合

记录当地历史发展轨迹的传统民间文化，是反映人文风格和特征的历史产物，具有重要的历史文化价值。在大数据的背景下，分析传统民俗文化的特征，挖掘传统民俗文化数据之间的内在联系和隐性信息，是传统文化信息科学领域的共同关注点。在对传统民间艺术的数字特征和大数据背景下的数据聚类方法进行研究的基础上，提出了一种基于动态自组织神经网络的深度自组织地图模糊C均值（DSOM-FCM）模型。网络，整合了陕西传统民俗文化资源。它不仅考虑了经济发展和精神文明建设的需要，还满足了陕西省社会文化发展的需要。首先，分析了民间传统艺术大数据的时空特征，然后将动态自组织神经网络的输入向量确定为6维属性数据。然后，在传统的自组织映射算法和模糊C均值技术的基础上，构建了基于DSOM-FCM的传统民间艺术资源整合模型。最后，使用陕西省传统文化数据集测试模型，实验结果如下：当SF = 0.35时，该算法的聚类数为3，实现了粗聚类。当SF = 0.7时，该算法的聚类数为6，实现了精细聚类。为了测试模型的效率和准确性，从分类误差，迭代时间，迭代次数和聚类数目的角度出发，进行了SOM和TreeGNG算法的比较实验。结果表明，本文设计和使用的算法在解决传统民俗文化的优化与融合方面效果良好。