Sogou New top 10 data mining and machine learning algorithms

One, the original 10 algorithms In 2006, IEEE's Data Mining Conference selected the top 10 algorithms: [see click on open link] Vector Machines C4.5k-MeansSupport (SVM) Maximization AprioriExpectation (EM) Neighbors PageRankAdaBoostk-Nearest (kNN) BayesClassification and Naive Regre...
read(1164) comment(0)

Sogou Machine learning algorithm series - the blog of the relevant machine learning algorithm directory

preface This part of the not to introduce what the specific machine learning algorithms, in front of the machine learning algorithm, I in the process of learning to look at others to write the material, but many authors write too difficult to understand, or is the release of the formula too much, so I think I to write to the point of this material can give everyone a reference to, of course, because of this talent and less learning, writing a blog or in the process of writing programs have any unreasonable or wrong place, trouble you a lot of pointed out that because of the value of your support to reflect my to do the work. Since the blog will be more and more, in this...
read(2254) comment(2)

Dirichlet Allocation Latent - Theory

introduction LDA (latent Dirichlet allocation) known as latent Dirichlet distribution, is a model for the semantic analysis of text is an important and at the same time, LDA models using Bayesian thinking some of the knowledge, the knowledge is the basis of statistical machine learning, these knowledge including function and the gamma distribution, beta function and distribution, function and the Dirichlet distribution, Bayes theorem, Gibbs sampling, and so on. In the next article, we introduce the core idea of LDA by the following aspects: Basic knowledge...
read(1972) comment(2)

Ad calculation - smooth CTR

First, the basic concept of advertising...
read(494) comment(1)

Scalable machine learning -- Classification -- Rate Prediction Click-through (click rate prediction)

Note: This is a study note, the record is a reference to the extension of the machine learning some of the contents of the English PPT visible reference links. This just my study notes, on the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope to contact me, if there is infringement, also want to inform, I will remove them as soon as possible. This part should be added to the experimental part, the part of the experiment in the latter part of the time to make up. The expansion of machine learning series mainly includes the following parts: an overview - Spark distributed processing - linear...
read(123) comment(0)

Scalable machine learning - gradient descent (Descent Gradient)

Note: This is a study note, the record is a reference to the extension of the machine learning some of the contents of the English PPT visible reference links. This just my study notes, on the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope to contact me, if there is infringement, also want to inform, I will remove them as soon as possible. This part should be added to the experimental part, the part of the experiment in the latter part of the time to make up. The expansion of machine learning series mainly includes the following parts: an overview - Spark distributed processing - linear...
read(112) comment(0)

Scalable machine learning - linear regression (Regression linear)

Note: This is a study note, the record is a reference to the extension of the machine learning some of the contents of the English PPT visible reference links. This just my study notes, on the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope to contact me, if there is infringement, also want to inform, I will remove them as soon as possible. This part should be added to the experimental part, the part of the experiment in the latter part of the time to make up. The expansion of machine learning series mainly includes the following parts: an overview - Spark distributed processing - linear...
read(143) comment(0)

Community division - Propagation Label

First, the summary of community division for the community, there is no a clear definition, the definition of a lot of community, such as community is refers to a group of nodes in a network, each other are similar, and other nodes within a group of nodes in the network is not similar. More general can be expressed as: community refers to the collection of nodes in the network, the internal connection of these nodes are more closely and the external connection is relatively sparse. Based on the above image representation, there are a lot of community partitioning algorithms, such as the Unfolding Fast algorithm introduced in the previous paper, Unfolding Fast algorithm is based on the module of the algorithm, module...
read(1044) comment(3)

Scalable machine learning -- Spark distributed processing

Note: This is a study note, the record is a reference to the extension of the machine learning some of the contents of the English PPT visible reference links. This just my study notes, on the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope to contact me, if there is infringement, also want to inform, I will remove them as soon as possible. The expansion of machine learning series mainly includes the following parts: summary Spark distributed processing Linear regression (Regression linear) Gradient descent (Gradien...
read(146) comment(0)

Scalable machine learning: an overview

Note: This is a study note, the record is a reference to the extension of the machine learning some of the contents of the English PPT visible reference links. This just my study notes, on the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope to contact me, if there is infringement, also want to inform, I will remove them as soon as possible. The expansion of machine learning series mainly includes the following parts: summary Spark distributed processing Linear regression (Regression linear) Gradient descent (Gradien...
read(87) comment(0)

Using Theano to understand deep learning - Encoder Auto

Note: this series is based on the contents of the reference and the collation, notes formed a series of depth study of the basic theory and practice materials, basic content and references consistent and on the topic name for "the Theano depth of understanding of learning" series, ruvin have any questions please consult. This article provides PDF version, welcome to obtain. "The Theano depth of understanding of the learning series is divided into 44 parts. This is the second part, in the first part of the algorithm is a supervised learning algorithm, in this section is mainly unsupervised learning algorithm and semi supervised learning algorithms, including...
read(143) comment(0)

Using Theano to understand deep learning - Neural Networks Convolutional

Note: this series is based on the contents of the reference and the collation, notes formed a series of depth study of the basic theory and practice materials, basic content and references consistent and on the topic name for "the Theano depth of understanding of learning" series, ruvin have any questions please consult. This article provides PDF version, welcome to obtain. "The use of Theano to understand deep learning" series is divided into 44 parts, including the first part of the main: Using Theano to understand deep learning - Regression Logistic Using Thea...
read(1182) comment(0)

UFLDL notes - depth network

Note: the recent plan to UFLDL tutorial to see once again, in fact, there are a lot of neural network and the depth of learning knowledge is useful, but only to learn deep learning some of the content is a bit redundant, so want to organize a notes, record neural network to study in depth some knowledge. The teaching material has been very good, online edition of the English version and translation of Chinese version, this is just my study notes, of the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope and I contact, if the content tort, also want to inform you that...
read(153) comment(0)

UFLDL notes: self learning

Note: the recent plan to UFLDL tutorial to see once again, in fact, there are a lot of neural network and the depth of learning knowledge is useful, but only to learn deep learning some of the content is a bit redundant, so want to organize a notes, record neural network to study in depth some knowledge. The teaching material has been very good, online edition of the English version and translation of Chinese version, this is just my study notes, of the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope and I contact, if the content tort, also want to inform you that...
read(230) comment(0)

UFLDL notes - Softmax regression

Note: the recent plan to UFLDL tutorial to see once again, in fact, there are a lot of neural network and the depth of learning knowledge is useful, but only to learn deep learning some of the content is a bit redundant, so want to organize a notes, record neural network to study in depth some knowledge. The teaching material has been very good, online edition of the English version and translation of Chinese version, this is just my study notes, of the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope and I contact, if the content tort, also want to inform you that...
read(234) comment(0)

UFLDL notes - sparse self encoder

Note: the recent plan to UFLDL tutorial to see once again, in fact, there are a lot of neural network and the depth of learning knowledge is useful, but only to learn deep learning some of the content is a bit redundant, so want to organize a notes, record neural network to study in depth some knowledge. The teaching material has been very good, online edition of the English version and translation of Chinese version, this is just my study notes, of the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope and I contact, if the content tort, also want to inform you that...
read(189) comment(0)

UFLDL notes - neural network

Note: the recent plan to UFLDL tutorial to see once again, in fact, there are a lot of neural network and the depth of learning knowledge is useful, but only to learn deep learning some of the content is a bit redundant, so want to organize a notes, record neural network to study in depth some knowledge. The teaching material has been very good, online edition of the English version and translation of Chinese version, this is just my study notes, of the contents of the original tutorial comb, also a reference to figure some original tutorial, if content have any mistake, hope and I contact, if the content tort, also want to inform you that...
read(331) comment(0)

Python skills - Python read files

In Python, read the file commands are as follows: three: Read () readLine () readlines () () () 1, read () Disposable read () function read, read is the entire contents of the file, and will be assigned to a string. Such as: Implementation results are: Note that in the read () function, you can specify the size of the read, such as read (5) The final result is: ...
read(134) comment(0)

Feature space in machine learning

Disclaimer: this blog is mainly the references in the PPT learning down some of the notes, sorted out and share with everyone, if notes have any errors also, please don't hesitate to pointed out that the text may use some plots of the original author, if the violation to the rights and interests of the author, please let us know, I will delete, thank you. First, the process of machine learning process application machine learning algorithm can be divided into: collecting data Data processing, feature extraction Training model Model deployment Application and feedback of the model The specific convergence relationship as shown below: two, the key problem of machine learning in machine learning is mainly as follows: three...
read(624) comment(0)

Python skills - list and string to each other

In Python programming, often involves the conversion between the string and the list problem, the following will be between the conversion to do a comb. 1, list converted into a string Command: list () example 2, the string is converted to list Command: ".Join" (list) Among them, the quote is the character of the segmentation, such as ",", ";", "\t" and so on. Example:...
read(146) comment(0)

Using Theano to understand deep learning - Perceptron Multilayer

, multilayer perceptron mlp1 and MLP overview for containing a single hidden layer of a multilayer perceptron (MLP) single-hidden-layer multi layer perceptron, can regard it as a special logistic regression classifier, the logistic regression classifier first through a nonlinear transformation phi \Phi (non-linear transformation of sample input nonlinear transformation, then will be transformed as the value of the status.
read(1580) comment(0)

Using Theano to understand deep learning - Regression Logistic

One, Regression1 LR, Logistic model Logistic regression is a generalized linear model, which belongs to a linear classification model, in its model there are two main parameters, namely: the weight matrix WW and the bias vector bb. In Logistic regression, the input vector is mapped to a set of hyper planes, each of which represents a class. The distance from the input vector to the hyper plane represents the probability that the input vector belongs to the member of the corresponding class. The input vector XX, the probability that it belong to the category of II is: P (Y=i, x, W, b) =sof...
read(2710) comment(0)
98 data a total of 5 pagesOne Two Three Four Five ... Next page Shadowe
    personal data
    • visit180267 times
    • Integral:Two thousand seven hundred and ninety-seven
    • Grade
    • Rank:7507th name
    • original95
    • Reproduced:1
    • Translation:0
    • Comments:171
    Contact me
    Email:zhaozhiyong1989@126.com

    Advertised

    In machine learning now basically divided into three generations: the first generation: non distributed; second generation tools such as the mahout and rapidminer implementation based on extension of Hadoop; the third generation, such as spark and storm to achieve real-time and iterative data processing

    Blog column
    Latest comments