逻辑回归数据_数据科学中的逻辑回归

逻辑回归数据

逻辑回归 (Logistic Regression)

Logistic regression is an applied mathematics analysis methodology accustomed to predict a data price supported previous observations of a data set. Logistic regression has become a very important tool within the discipline of machine learning. The approach permits associate degree formula being employed in a very machine learning application to classify incoming data supported historical data. As additional relevant data comes in, the formula ought to get well at predicting classifications inside data sets.
Logistic regression can even play a job in data preparation activities by permitting data sets to be placed into specifically predefined buckets throughout the extract, transform, load (ETL) method to stage the data for analysis.

逻辑回归是一种应用数学分析方法，习惯于预测支持数据集先前观察的数据价格。逻辑回归已成为机器学习领域中非常重要的工具。该方法允许在非常机器学习的应用程序中使用副学历公式来对输入数据支持的历史数据进行分类。随着更多相关数据的涌入，该公式应能很好地预测数据集中的分类。
通过允许将数据集放置在整个提取，转换，加载(ETL)方法中的特定预定义存储桶中，以进行数据分析，逻辑回归甚至可以在数据准备活动中发挥作用。

A logistics regression model predicts a dependent data variable by analyzing the link between one or additional existing freelance variables. For instance, logistic regression may be accustomed to predict whether or not a political candidate can win or lose an associate degree election or whether a high school student is admitted to a specific faculty.

物流回归模型通过分析一个或其他现有自由职业变量之间的联系来预测因变量。例如，逻辑回归可以习惯于预测政治候选人是否可以赢得或失去副学士学位选举，或者高中生是否被录取到特定的系。

The ensuing analytical model will take into thought multiple input criteria. Within the case of faculty acceptance, the model may contemplate factors like the student’s grade average, Sabbatum score and variety of extracurricular activities. Supported historical data concerning earlier outcomes involving equivalent input criteria, it then scores new cases on their chance of falling into a specific outcome class.

随后的分析模型将考虑多种输入标准。在教师接受的情况下，该模型可以考虑诸如学生的平均成绩，Sabbatum分数和各种课外活动之类的因素。支持的历史数据涉及涉及等效输入标准的早期结果，然后根据新案例落入特定结果类别的机会对新案例进行评分。

物流回归的目的和样本 (Purpose and samples of logistics regression)

Logistic regression is one among the foremost unremarkably used machine learning algorithms for binary classification issues, that are issues with 2 category values, as well as predictions like "this or that", "yes or no" and "A or B".

Logistic回归是用于二进制分类问题的最重要的机器学习算法之一，它是具有2个类别值的问题，以及诸如“ this or that” ， “ yes or no”和“ A或B”之类的预测。

The purpose of logistics regression is to estimate the possibilities of events, as well as crucial a relationship between options and therefore the chances of specific outcomes.
On examination of this is often predicting if a student can pass or fail associate degree exam once the quantity of hours spent finding out is provided as a feature and therefore the variables for the response have 2 values: pass and fail.

后勤回归的目的是估计事件的可能性，以及期权之间的关系，以及由此得出的特定结果的可能性。
在检查时，通常会预测学生是否可以通过一次发现花费的小时数作为一项功能来通过或通过副学士学位考试，因此答案的变量具有两个值：通过和失败。

Organizations will use insights from logistic regression outputs to reinforce their business ways so that they can do their business goals, as well as reducing expenses or losses and increasing ROI in promoting campaigns, for instance.

组织将利用逻辑回归输出的见解来加强其业务方式，以便他们能够实现其业务目标，并减少开支或损失并提高促销活动中的投资回报率。

An e-commerce company that mails expensive promotional offers to clients would like to understand whether or not a specific customer is probably going to retort to the offers or not. For instance, they'll wish to understand whether or not that client is a "responder" or a "non-answerer." In promoting, this is often referred to as propensity to answer modeling.

一家向客户邮寄昂贵促销优惠的电子商务公司希望了解特定客户是否可能会拒绝这些优惠。例如，他们希望了解该客户是“响应者”还是“非回答者”。在推广中，这通常被称为回答建模的倾向。

Likewise, a Mastercard company develops a model to decide whether or not to issue a credit card to a client or not an attempt to predict whether the customer goes to default or not on the credit card supported such characteristics as annual financial gain, monthly Mastercard payments, and variety of defaults. In banking idiom, this is often called default propensity modeling.

同样，万事达卡公司开发一种模型，以决定是否向客户发行信用卡，以尝试预测客户是否因支持的信用卡而违约，例如年度财务收益，每月万事达卡付款以及各种默认值。在银行习惯用法中，这通常称为默认倾向建模。

物流回归的用途 (Uses of Logistics Regression)

Logistic regression has become significantly widespread in on-line advertising, facultative marketers to predict the chance of specific web site users UN agency can click on specific advertisements as an affirmative or no proportion.

逻辑回归在在线广告中已变得非常普遍，兼职营销人员可以预测特定网站用户的机会。联合国机构可以按肯定比例或不按比例点击特定广告。

Logistic regression can even be used in:

逻辑回归甚至可以用于：

Healthcare to spot risk factors for diseases and set up preventive measures.
医疗保健要发现疾病的危险因素并制定预防措施。
Weather forecasting apps to predict downfall and climate.
天气预报应用程序可预测降雨和气候。
Voting apps to see if voters can vote for a specific candidate.
投票应用程序可查看选民是否可以为特定候选人投票。
Insurance to predict the probabilities that a policyholder can die before the term of the policy expires supported bound criteria, like gender, age, and physical examination.
保险以预测保单持有人在保单期限届满之前可能死亡的可能性，这是受支持的受约束标准，例如性别，年龄和身体检查。
Banking to predict the probabilities that a loan mortal can default a loan or not, supported annual financial gain, past defaults, and past debts.
银行业务预测贷款人可以违约或不违约的概率，支持的年度财务收益，过去的违约和过去的债务。

Logistic回归与统计回归 (Logistic Regression vs. Statistical Regression)

The main distinction between logistics regression and statistical regression is that logistic regression provides a relentless output, whereas statistical regression provides never-ending output.

后勤回归和统计回归之间的主要区别是逻辑回归提供了无休止的输出，而统计回归提供了无休止的输出。

In logistics regression, the end result, like a variable, solely features a restricted variety of attainable values. However, in statistical regression, the end result is continuous, which implies that it will have anybody of an infinite variety of attainable values.

在物流回归中 ，最终结果像变量一样，仅具有有限的各种可获得值。但是，在统计回归中，最终结果是连续的，这意味着它将拥有无限多种可实现值的任何人。

Logistic regression is employed once the response variable is categorical, like yes/no, true/false and pass/fail. statistical regression is employed once the response variable is continuous, like a variety of hours, height and weight.

一旦响应变量是分类的(例如是/否，是/否和通过/失败)，就采用逻辑回归 。一旦响应变量是连续的(如各种小时，身高和体重)，就采用统计回归。

For example, given data on the time a student spent finding out which student's examination scores, logistics regression, and statistical regression will predict various things.

例如，给定有关学生花费时间的数据，以找出哪些学生的考试成绩，后勤回归和统计回归将预测各种情况。

With logistics regression predictions, solely specific values or classes are allowed. Therefore, logistics regression will predict whether or not the coed passed or failing. Since statistical regression predictions are continuous, like numbers vary, it will predict the student's take a look at the score on a scale of 0-100.

对于物流回归预测，仅允许使用特定值或类别。因此，物流回归将预测男女学生是否通过或失败。由于统计回归预测是连续的，就像数字变化一样，它将预测学生以0-100的比例查看分数。