...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>SN Applied Sciences >IoT‑based group size prediction and recommendation system using machine learning and deep learning techniques
【24h】

IoT‑based group size prediction and recommendation system using machine learning and deep learning techniques

机译：基于机器学习和深度学习技术的基于IOT的组大小预测和推荐系统

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In an open source software development environment, it is hard to decide the number of group members required forresolving software issues. Developers generally reply to issues based totally on their domain knowledge and interest,and there are no predetermined groups. The developers openly collaborate on resolving the issues based on many factors,such as their interest, domain expertise, and availability. This study compares eight different algorithms employingmachine learning and deep learning, namely-Convolutional Neural Network, Multilayer Perceptron, Classification andRegression Trees, Generalized Linear Model, Bayesian Additive Regression Trees, Gaussian Process, Random Forest andConditional Inference Tree for predicting group size in five open source software projects developed and managedusing an open source development framework GitHub. The social information foraging model has also been extendedto predict group size in software issues, and its results compared to those obtained using machine learning and deeplearning algorithms. The prediction results suggest that deep learning and machine learning models predict better thanthe extended social information foraging model, while the best-ranked model is a deep multilayer perceptron((R.M.S.E.sequelize-1.21, opencv-1.17, bitcoin-1.05, aseprite-1.01, electron-1.16). Also it was observed that issue labelshelped improve the prediction performance of the machine learning and deep learning models. The prediction resultsof these models have been used to build an Issue Group Recommendation System as an Internet of Things applicationthat recommends and alerts additional developers to help resolve an open issue.
机译：在开源软件开发环境中，很难确定所需的组成员数量解决软件问题。开发人员通常回复基于域名知识和兴趣的问题，并且没有预定的群体。开发商公开合作基于许多因素解决问题，如他们的兴趣，域名专业知识和可用性。本研究比较了八种不同的算法机器学习和深度学习，即卷积神经网络，多层默认，分类和回归树，广义线性模型，贝叶斯添加剂回归树，高斯过程，随机森林和用于预测组大小的条件推理树，在开发和管理的五个开源软件项目中使用开源开发框架GitHub。社会信息觅食模式也已经扩展为了预测软件问题中的群组大小，与使用机器学习和深度获得的结果相比学习算法。预测结果表明，深度学习和机器学习模型预测比扩展的社交信息觅食模型，而最佳排名的模型是深层多层的感知者（（r.m.s.e。后源-1.21，OpenCV-1.17，比特键-1.05，Aseprite-1.01，Electron-1.16）。也观察到问题标签帮助改善机器学习的预测性能和深度学习模型。预测结果这些模型已被用于构建一个问题组推荐系统作为事物的互联网这推荐并提醒其他开发人员帮助解决开放问题。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号