基于神经网络词嵌入的知识共享比较研究
首发时间:2023-02-09
摘要:[目的]对我国学术平台和社会化问答平台的知识共享进行比较,旨在推动我国知识共享研究取得更多的成果。[方法]word2vec是新兴的神经网络词嵌入算法,不仅计算成本低,而且准确度高,能够同时在语法层面和语义层面对词语的相似度进行有效的测度。[结果]先结合"与知识共享和知识分享最相似的前20个词语"和"与知识最相似的前20个词语"对相似词语进行了比较,然后,利用降维和可视化方法对词语的词嵌入概貌进行了比较,研究结果展现了我国学术平台和社会化问答平台在知识共享方面的差异。[结论]本文创新性地利用神经网络词嵌入算法对我国学术平台和社会化问答平台的知识共享进行了比较分析,为知识共享研究提供了新的视角。
关键词: 知识共享 word2vec 学术平台 社会化问答平台 比较研究
For information in English, please click here
Comparative Study on Knowledge Sharing Based on Neural Network Word Embedding
Abstract:[Objective] This paper compares knowledge sharing between academic and social Q&A platforms in China, with the aim of facilitating the progress of knowledge sharing research in China. [Methods] word2vec is a new neural network word embedding algorithm. It has low computation cost and high accuracy, being able to effectively measure similarity of words at both grammatical and semantic levels. [Results] Based on the top 20 words which are most similar to knowledge sharing and knowledge, comparisons of similar words were conducted. Then, employing dimension reduction and visualization algorithm to conduct comparison of general word pictures of word embedding. The results illustrate the differences of knowledge sharing between academic and social Q&A platforms. [Conclusions] This paper innovatively employs the neural network word embedding algorithm to compare knowledge sharing on academic and social Q&A platforms, providing a new perspective for knowledge sharing research.
Keywords: knowledge sharing word2vec academic platform social Q&A platform comparative study
引用
No.****
动态公开评议
共计0人参与
勘误表
基于神经网络词嵌入的知识共享比较研究
评论
全部评论