When we have to predict the value of a [categorical](https://en.wikipedia.org/wiki/Categorical_variable) (or discrete) outcome we use [logistic regression](https://en.wikipedia.org/wiki/Logistic_regression). I believe we use [linear regression](https://en.wikipedia.org/wiki/Linear_regression) to also predict the value of an outcome given the input values.

Then, what is the difference between the two methodologies?

What is the difference between linear regression and logistic regression?

[Cosine Similarity article on Wikipedia](http://en.wikipedia.org/wiki/Cosine_similarity)

Can you show the vectors here (in a list or something)
and then do the math, and let us see how it works?



Can someone give an example of cosine similarity, in a very simple, graphical way?

Can someone explain what the difference is between classification and clustering in data mining?

If you can, please give examples of both to understand the main idea.


Difference between classification and clustering in data mining?

What technology goes in behind the screens of Amazon recommendation technology? I believe that Amazon recommendation is currently the best in the market, but how do they provide us with such relevant recommendations?

Recently, we have been involved with similar recommendation kind of project, but would surely like to know about the in and outs of the Amazon recommendation technology from a technical standpoint. 

Any inputs would be highly appreciated. 

**Update:**

This [patent][1] explains how personalized recommendations are done but it is not very technical, and so it would be really nice if some insights could be provided. 

From the comments of Dave, [Affinity Analysis][2] forms the basis for such kind of Recommendation Engines. Also here are some good reads on the Topic

 1. [Demystifying Market Basket Analysis][3]
 2. [Market Basket Analysis][4]
 3. [Affinity Analysis][5]

**Suggested Reading:**

 3. [Data Mining: Concepts and Technique][6]


  [1]: http://www.google.com/patents?id=dtp6AAAAEBAJ&amp;printsec=abstract&amp;zoom=4&amp;source=gbs_overview_r&amp;cad=0#v=onepage&amp;q=&amp;f=false
  [2]: http://en.wikipedia.org/wiki/Affinity_analysis
  [3]: http://www.information-management.com/specialreports/20061031/1067598-1.html
  [4]: http://people.revoledu.com/kardi/tutorial/MarketBasket/index.htm
  [5]: http://it.toolbox.com/blogs/enterprise-solutions/affinity-analysis-14726
  [6]: http://www.amazon.com/Data-Mining-Concepts-Techniques-Management/dp/1558604898

How does the Amazon Recommendation feature work?

I have noticed that when One Hot encoding is used on a particular data set (a matrix) and used as training data for learning algorithms, it gives significantly better results with respect to prediction accuracy, compared to using the original matrix itself as training data. How does this performance increase happen? 

Why does one hot encoding improve machine learning performance?

When we calculate the F-Measure considering both Precision and Recall, we take the harmonic mean of the two measures instead of a simple arithmetic mean. 

What is the intuitive reason behind taking the harmonic mean and not a simple average?

Why is the F-Measure a harmonic mean and not an arithmetic mean of the Precision and Recall measures?

Expectation Maximization (EM) is a kind of probabilistic method to classify data. Please correct me if I am wrong if it is not a classifier. 

What is an intuitive explanation of this EM technique? What is `expectation` here and what is being `maximized`?

What is an intuitive explanation of the Expectation Maximization technique?

So let&#39;s say I have an array like this:

    [1,1,2,3,10,11,13,67,71]

Is there a convenient way to partition the array into something like this?

    [[1,1,2,3],[10,11,13],[67,71]]

I looked through similar questions yet most people suggested using k-means to cluster points, like [scipy][1], which is quite confusing to use for a beginner like me. Also I think that k-means is more suitable for two or more dimensional clustering right? Are there any ways to partition an array of N numbers to many partitions/clustering depending on the numbers?

Some people also suggest rigid range partitioning, but it doesn&#39;t always render the results as
 expected

  [1]: http://docs.scipy.org/doc/scipy/reference/cluster.vq.html#

1D Number Array Clustering

I&#39;m using scikit-learn in Python to develop a classification algorithm to predict the gender of certain customers. Amongst others, I want to use the Naive Bayes classifier but my problem is that I have a mix of categorical data (ex: &quot;Registered online&quot;, &quot;Accepts email notifications&quot; etc) and continuous data (ex: &quot;Age&quot;, &quot;Length of membership&quot; etc). I haven&#39;t used scikit much before but I suppose that that Gaussian Naive Bayes is suitable for continuous data and that Bernoulli Naive Bayes can be used for categorical data. However, since I want to have **both** categorical and continuous data in my model, I don&#39;t really know how to handle this. Any ideas would be much appreciated!

Mixing categorial and continuous data in Naive Bayes classifier using scikit-learn

I understand what Gradient Descent does. Basically it tries to move towards the local optimal solution by slowly moving down the curve. I am trying to understand what is the actual difference between the plan gradient descent and the newton&#39;s method?

From Wikipedia, I read this short line &quot;Newton&#39;s method uses curvature information to take a more direct route.&quot; What does this intuitively mean?



What is the difference between Gradient Descent and Newton&#39;s Gradient Descent?

I&#39;m trying to compile a list of information that is accessible via javascript such as:

 - Geo-location
 - IP address
 - Browser software
 - Exit location
 - Entrance location

I understand that a user can alter any of this information and that it&#39;s reliability is purely trust related, but I am still interested in what other information can be mined from the client.

What information can we access from the client?

Given a vector of scores and a vector of actual class labels, how do you calculate a single-number AUC metric for a binary classifier in the R language or in simple English? 

Page 9 of &lt;a href=&quot;http://www.site.uottawa.ca/~stan/csi7162/presentations/William-presentation.pdf&quot;&gt;&quot;AUC: a Better Measure...&quot;&lt;/a&gt; seems to require knowing the class labels, and here is &lt;a href=&quot;http://dwinnell.com/SampleError.m&quot;&gt;an example in MATLAB&lt;/a&gt; where I don&#39;t understand 

    R(Actual == 1))

Because R (not to be confused with the R language) is defined a vector but used as a function?

Calculate AUC in R?

I am doing some research about different data mining techniques and came across something that I could not figure out.
If any one have any idea that would be great.

In which cases is it better to use a Decision tree and other cases a Naive Bayes classifier?

Why use one of them in certain cases? And the other in different cases?
(By looking at its functionality, not at the algorithm)

Anyone have some explanations or references about this?


Decision tree vs. Naive Bayes classifier

In my understanding, I thought PCA can be performed only for continuous features. But while trying to understand the difference between onehot encoding and label encoding came through a post in the following link:

[When to use One Hot Encoding vs LabelEncoder vs DictVectorizor?](https://datascience.stackexchange.com/questions/9443/when-to-use-one-hot-encoding-vs-labelencoder-vs-dictvectorizor)

It states that one hot encoding followed by PCA is a very good method, which basically means PCA is applied for categorical features.
Hence confused, please suggest me on the same.

All Data Mining Solutions on Gang of Coders

Total of 14 Data Mining Solutions