SVM, Perceptron, LDA, Logistic Regression etc.

本文对比了机器学习中的SVM、感知机、LDA和逻辑回归,探讨了它们之间的关系和区别。内容包括Fisher线性判别分析与逻辑回归的联系,SVM与感知机的优化方法和适用场景,以及SVM与逻辑回归的差异。此外,还提到了在高维空间中SVM的优势以及在不同问题类型上的适用性选择。
摘要由CSDN通过智能技术生成

机器学习常见算法分类汇总

Are Fisher’s linear discriminant and logistic regression classifier related?

The Efficiency of Logistic Regression Compared to Normal Discriminant Analysis
Bradley Efron
Journal of the American Statistical Association
Vol. 70, No. 352 (Dec., 1975), pp. 892-898

What is the difference between the perceptron learning algorithm and SVM?

Difference between a SVM and a perceptron

The Perceptron does not try to optimize the separation "distance". As long as it finds a hyperplane that separates the two sets, it is good. SVM on the other hand tries to maximize the "support vector", i.e., the distance between two closest opposite sample points.

The SVM typically tries to use a "kernel function" to project the sample points to high dimension space to make them linearly separable, while the perceptron assumes the sample points are linearly separable.
### sklearn支持的机器学习模型 #### 一、监督学习模型 supervised learning models are used when the training data includes both input features and output labels, with the goal of predicting accurate outcomes for new inputs. - **线性模型** - `LinearRegression`用于回归分析,通过拟合最小二乘法来建立输入特征与连续型目标变量之间的关系[^1]。 - `LogisticRegression`适用于分类任务,尽管名称中有“回归”,但实际上是一种广义线性分类器,能够处理二元或多类别分类问题[^2]。 - **支持向量机(SVM)** - SVMs can perform binary classification as well as regression tasks using different kernel functions like linear, polynomial or RBF. - **决策树(Decision Trees)** - Decision trees learn simple decision rules from data to predict target values. They support both numerical and categorical data without requiring much preprocessing work on part of users[^3]. - **集成方法(Ensemble Methods)** - Ensemble methods combine multiple weak learners into a strong learner which generally gives better performance than individual ones alone. - `RandomForestClassifier`: A type of ensemble method that constructs many decision trees at training time and outputs class that is mode among them during prediction phase. - `GradientBoostingClassifier/Regressor`: Another popular boosting algorithm where each subsequent tree tries correcting errors made by previous members within sequence until satisfactory accuracy level achieved over entire dataset provided for fitting process[^4]. - **神经网络(Neural Networks)** - Multi-layer Perceptron (MLP): This module implements feed-forward artificial neural networks trained via backpropagation either supervisedly against known classes/categories or unsupervised manner depending upon application requirements set forth beforehand. #### 二、无监督学习模型 Unsupervised algorithms aim to find hidden patterns or intrinsic structures present inside datasets having only feature vectors but no corresponding label information available alongside it whatsoever. - **聚类(Clustering)** - K-means clustering partitions n observations into k clusters wherein each observation belongs exactly one cluster whose centroid has nearest mean value compared others around itself; this approach works best under assumption spherical shaped groups exist naturally within multidimensional space occupied points being analyzed here. - **降维(Dimensionality Reduction)** - Principal Component Analysis (PCA), Linear Discriminant Analysis(LDA), etc., these techniques help reduce dimension while preserving most variance possible across transformed components so they remain representative enough about original distribution properties after transformation applied onto raw measurements collected initially before analysis started off course. ```python from sklearn.linear_model import LogisticRegression from sklearn.svm import SVC from sklearn.tree import DecisionTreeClassifier from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier from sklearn.neural_network import MLPClassifier from sklearn.cluster import KMeans from sklearn.decomposition import PCA ```
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值