Linear regression is probably the most popular machine learning algorithm. Linear regression is to find a straight line and make it fit the data points in the scatter plot as much as possible. It tries to represent the independent variable (x? Value) and numerical results (y? Value). Then you can use this line to predict future values!
Logical regression is similar to linear regression, but it is used when the output is binary (that is, the result can only have two possible values). Is the prediction of the final output nonlinear? s？ Type function, called. Logical function, g ().
Decision trees can be used for regression and classification tasks.
Naive Bayes is based on Bayes theorem. It measures the probability of each class, and the conditional probability of each class is given by x? The value. This algorithm is used to classify problems and get a binary "yes?" /? The result of "no" Look at the equation below.
Support Vector Machine (SVM) is a supervised algorithm for classification problems. Support vector machine tries to draw two lines between data points, and the distance between them is the largest. To this end, we draw the data item as? n？ A point in a dimensional space, where n? Is the number of input elements. On this basis, support vector machine finds an optimal boundary, called hyperplane, and optimally separates possible outputs through class labels.
K-？ K nearest neighbor (KNN) algorithm is very simple. KNN？ By searching the entire training set? k？ The most similar example, namely? k？ Neighbor, for all this? k？ An instance is assigned a common output variable to classify objects.
K-？ K-means clusters by classifying data sets. For example, the algorithm can be used to group users according to their purchase history. Found it in the dataset? k？ A bunch. K-？ Mean is used for unsupervised learning, so we only need to use training data? X, and the number of clusters we want to identify? k .
Random forest is a very popular integrated machine learning algorithm. The basic idea of this algorithm is that many people's opinions are more accurate than personal opinions. In random forests, we use decision tree integration (see decision tree).
Because we can capture a lot of data today, the problem of machine learning becomes more complicated. This means that training is extremely slow and it is difficult to find a good solution. This problem is often called "dimension disaster".
Artificial Neural Network (ANN) can handle large and complex machine learning tasks. Neural network is essentially a set of interconnected layers composed of weighted edges and nodes, called neurons. Between the input layer and the output layer, we can insert multiple hidden layers. Artificial neural networks use two hidden layers. In addition, deep learning also needs to be dealt with.