A machine, learning algorithm that incorrectly classifies new data values, may have a problem with population imbalances in:机器学习算法对新数据值进行错误分类,由于_______,可能存在模型偏移的问题::()?
(A)Model training data模型训练数据
(B)Metadata management元数据管理
(C)Reference data management参考数据管理
(D)Big data architecture大数据架构
(E)Predictive analytics预测分析 答案解析: 一般来说,机器学习结果不准,要么就是算法选择或调参有问题,要么就是训练的样本有问题该题并未提到算法选择或调参,那么必定是训练的样本数据有误 ~ P748,下面内容只作参考,以老师解释为准,该题是超纲题目
参考答案
继续答题:下一题
更多CDMP数据治理试题
- 1Information gaps-the difference between what information is needed and whatever trustworthy information is currently available-represent信息差距——所需信息与当前可用的可信信息之间的差异——表示:()?
- 2A data lineage tool enables a user to:数据血缘工具使用户能够::()?
- 3The key architecture domains include:关键架构领域包括::()?
- 4When outsourcing information management functions, organisations can..当外包信息管理系统时,企业可以::()?
- 5Common understanding of the core business concepts and terminology is the objective of which deliverable?对核心业务概念和术语的共同理解是哪个交付件的目标?:()?
- 6According to the ISO/IEC 42010:2007 Software and Systems Engineering-Architecture Description, which of the following describes the definition of architecture:根据ISO/IEC 42010:2007软件和系统工程-体系结构描述,以下哪一项描述了体系结构的定义::()?