%0 Journal Article
%T 基于卷积神经网络的教育众筹成败预测<br>Prediction of Educational Crowds’ Success or Failure Based on Convolutional Neural Network
%A 骆丽萍
%A 黄洁
%A 张雅歌
%J Software Engineering and Applications
%P 319-325
%@ 2325-2278
%D 2019
%I Hans Publishing
%R 10.12677/SEA.2019.86039
%X 教育众筹在一定程度上可以优化、检验教育课程，有效整合社会资源及教育资源，为地方缓解资金压力，如果众筹失败，将造成巨大的时间成本，因此对众筹项目结果进行预测研究具有重要意义[1]。本文针对教育众筹的成败预测问题，将卷积神经网络模型运用于教育众筹成败预测中，通过采用神经网络语言模型word2vec对文本进行词向量的训练，并用训练好的词向量表示文本，使用卷积神经网络对文本进行抽象特征的提取，在提取出抽象特征的基础上，对纯文本信息进行网络训练和预测，获得了88.16%的测试正确率。<br />
Educational crowdsourcing, to a certain extent, can optimize and test educational courses, effec-tively integrate social resources and educational resources, and alleviate the financial pressure for local governments. If crowdsourcing fails, it will cause huge time costs. Therefore, it is of great significance to predict the results of crowdsourcing projects [1]. Aiming at the problem of predicting the success or failure of educational crowdsourcing, the convolutional neural network model is applied to predict the success or failure of educational crowdsourcing. Word2vec, a language model of neural network, is used to train the word vectors of the text, and the trained word vectors are used to represent the text. The abstract features of the text are extracted by using convolutional neural network. On the basis of extracting abstract features, network training and prediction of text information are carried out, and 88.16% of the test accuracy is obtained.
%K 教育众筹，成败预测，卷积神经网络<br>Educational Crowdfunding
%K Predicting the Success or Failure
%K Convolutional Neural Network
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=33263