论文部分内容阅读
【目的】基于相关性、多样性原则,利用图片推荐的方法刺激群体创新。【应用背景】在群体创意的环境中提供刺激信息刺激群体思维。【方法】基于文本分词、网页分析以及余弦相似度算法,建立图片推荐系统的模型,提出最大差异化算法,向正在创意的群体推荐多样性高的图片信息。【结果】通过实验研究,证明采用最大差异化算法的图片推荐系统对于群体创新绩效的促进作用。【局限】图片最大差异化算法中的相关性和差异性主要基于图片的描述信息,与图片本身的内容有一定的差异,因此基于图片描述信息的最大差异化算法存在局限性。【结论】在群体创新过程中,通过不断推荐差异化图片的方法能够提高群体创意的绩效。
【Objective】 Based on the principle of relevance and diversity, we use group recommendation to stimulate group innovation. 【Application Background】 In the group creative environment to provide stimulating information to stimulate group thinking. 【Method】 Based on text word segmentation, webpage analysis and cosine similarity algorithm, a model of image recommendation system was established and the maximum difference algorithm was proposed to recommend highly diverse image information to creative groups. 【Result】 Through experimental research, we prove that the image recommendation system with the largest difference algorithm can promote the innovation performance of the group. Limitations The correlation and differences in the maximum difference algorithm are mainly based on the description information of the image, which is different from the content of the image itself. Therefore, the maximum difference algorithm based on the image description information has limitations. 【Conclusion】 In the process of group innovation, the performance of group creativity can be improved by constantly recommending differentiated pictures.