Module 02 · From Understanding to Preparation and From Modeling to Evaluation

Glossary: From Modeling to Evaluation术语表：从建模到评估

Welcome! This alphabetized glossary contains many of the terms you'll find within this lesson. These terms are important for you to recognize when working in the industry, when participating in user groups, and when participating in other certificate programs.
欢迎！这个按字母顺序排列的术语表包含了本课程中许多你将遇到的术语。这些术语对于你在行业内工作、参与用户组以及参与其他证书项目时识别它们非常重要。

Term 术语	Definition 定义
Binary classification model 二分类模型	A model that classifies data into two categories, such as yes/no or stop/go outcomes. 一种将数据分为两类（如“是/否”或“停止/进行”）的模型。
Data compilation 数据编译	The process of gathering and organizing data required for modeling. 为建模而收集和组织数据的过程。
Data modeling 数据建模	The stage in the data science methodology where data scientists develop models, either descriptive or predictive, to answer specific questions. 数据科学方法论中，数据科学家开发描述性或预测性模型以回答特定问题的阶段。
Descriptive model 描述性模型	A type of model that examines relationships between variables and makes inferences based on observed patterns. 一种研究变量间关系并根据观察到的模式进行推断的模型类型。
Diagnostic measure based tuning 基于诊断措施的调优	The process of fine-tuning the model by adjusting parameters based on diagnostic measures and performance indicators. 通过调整参数，根据诊断措施和性能指标对模型进行微调的过程。
Diagnostic measures 诊断措施	The evaluation of a model's performance of a model to ensure that the model functions as intended. 评估模型性能，以确保模型按预期运行。
Discrimination criterion 区分标准	A measure used to evaluate the performance of the model in classifying different outcomes. 用于评估模型在分类不同结果方面的性能的指标。
False-positive rate 假阳性率	The rate at which the model incorrectly identifies negative outcomes as positive. 模型将负结果错误地识别为正结果的比率。
Histogram 直方图	A graphical representation of the distribution of a dataset, where the data is divided into intervals or bins, and the height of each bar represents the frequency or count of data points falling within that interval. 一种数据集分布的图形表示，其中数据被划分为区间或箱，每个条形的高度代表落入该区间内的数据点的频率或计数。
Maximum separation 最大分离	The point where the ROC curve provides the best discrimination between true-positive and false-positive rates, indicating the most effective model. ROC 曲线提供最佳真阳性率和假阳性率之间区分的点，表明最有效的模型。
Model evaluation 模型评估	The process of assessing the quality and relevance of the model before deployment. 在部署前评估模型质量和相关性的过程。
Optimal model 最优模型	The model that provides the maximum separation between the ROC curve and the baseline, indicating higher accuracy and effectiveness. ROC 曲线与基线之间分离度最大的模型，表明其具有更高的准确性和有效性。
Receiver Operating Characteristic (ROC) 受试者工作特征 (ROC)	Originally developed for military radar, the military used this statistical curve to assess the performance of binary classification models. 最初为军事雷达开发，军方使用该统计曲线来评估二元分类模型的性能。
Relative misclassification cost 相对误分类成本	This measurement is a parameter in model building used to tune the trade-off between true-positive and false-positive rates. 该度量是模型构建中的一个参数，用于调整真阳性率和假阳性率之间的权衡。
ROC curve (Receiver Operating Characteristic curve) ROC curve (受试者工作特征曲线)	A diagnostic tool used to determine the optimal classification model's performance. 一种用于确定最佳分类模型性能的诊断工具。
Separation 分离	Separation is the degree of discrimination achieved by the model in correctly classifying outcomes. 分离是指模型在正确分类结果时所达到的区分程度。
Statistical significance testing Statistical significance 测试	Evaluation technique to verify that data is appropriately handled and interpreted within the model. 一种评估技术，用于验证数据在模型中是否得到适当处理和解释。
True-positive rate 真阳性率	The rate at which the model correctly identifies positive outcomes. 模型正确识别正面结果的比率。