DataGenX - Atul's Blog: Analysis

Showing posts with label Analysis. Show all posts

Tuesday, 9 May 2017

Measuring Data Similarity or Dissimilarity #2

by Unknown / in Analysis, Analytics, data, Data Mining, Dissimilarity, Measures, Similarity /

Continuing from our last discussion 'Measuring Data Similarity or Dissimilarity #1', In this post we are going to see how to calculate the similarity or dissimilarity between Numeric Data Types.

2. For Numeric Attribute:

For measuring the dissimilarity between two numeric data points, the easiest or most used way to calculate the 'Euclidean distance', Higher the value of distance, higher the dissimilarity.
There are two more distance measuring methods named 'Manhattan distance' and 'Minkowski distance'. We are going to look into these one by one.

a. Euclidean distance:

Euclidean distance is widely used to calculate the dissimilarity between numeric data points, this is actually derived from 'Pythagoras Theorem' so also known as 'Pythagorean metric' or `L^2` norm.

Euclidean distance between two points `p(x_1, y_1)` and `q(x_2, y_2)` is the length which connects point p from point q.

`dis(p,q) = dis(q,p) = \sqrt((x_2 - x_1)^2 + (y_2 - y_1)^2) = \sqrt(\sum_(i=1)^N(q_i - p_i)^2)`

In One Dimention:

`dis(p,q) = dis(q,p) = \sqrt((q - p)^2) = q - p`

In Two Dimentions:

`dis(p,q) = dis(q,p) = \sqrt((q_1 - p_1)^2 + (q_2 - p_2)^2)`

In Three Dimentions:

`dis(p,q) = dis(q,p) = \sqrt((q_1 - p_1)^2 + (q_2 - p_2)^2 + (q_3 - p_3)^2)`

In N Dimentions:

`dis(p,q) = dis(q,p) = \sqrt((q_1 - p_1)^2 + (q_2 - p_2)^2 + (q_3 - p_3)^2 +.......................+ (q_N - p_N)^2)`

b. Manhattan distance:

It is also known as "City Block" distance as it is calculated same as we calculate the distance between any two block of city. It is simple difference between the data points.

`dis(p, q) = |(x_2 - x_1)| + |(y_2 - y_1)| = \sum_(i=1)^N|(q_i - p_i)|`

Manhattan distance is also know as `L^1` norm.

c. Minkowski distance:

This is the generalized form of Euclidean or Manhattan distance and represented as -

`dis(p,q) = dis(q,p) = [(x_2 - x_1)^n + (y_2 - y_1)^n]^{1/n} = [\sum_(i=1)^N(q_i - p_i)^n]^{1/n}`

where n = 1, 2, 3.......

Like the below page to get update
https://www.facebook.com/datastage4you
https://twitter.com/datagenx
https://plus.google.com/+AtulSingh0/posts
https://datagenx.slack.com/messages/datascience/

Measuring Data Similarity or Dissimilarity #1

by Unknown / in Analysis, data, Data Mining, Dissimilarity, Graphical, Plotting, Similarity, Statistics /

Yet another question is in data mining to measure whether two datasets are similar or not. There are so many ways to calculate these values based on Data Type. Let's see into these methods -

1. For Binary Attribute:

Binary attributes are those which is having only two states 0 or 1, where 0 means attribute is absent and 1 means it is present. For calculating similarity/dissimilarity between binary attributes we use contingency table -

q - if i and j both are equal to 1
r - if i is 1 and j is 0
s - if i is 0 and j is 1
t - if i and j both are equal to 0
p - total ( q+r+s+t)

a. Symmetric Binary Dissimilarity -

For symmetric binary attribute, each state is equally valuable. If i and j are symmetric binary attribute then dissimilarity is calculates as -

` d(i, j) = \frac{r + s}{q + r + s + t} `

b. Asymmetric Binary Dissimilarity -

For asymmetric binary attribute, two states are not equally important. Any one state overshadow the other, such binary attribute are often called "monary" (having one state). For these kind of attribute, dissimilarity is calculates as -

`d(i, j) = \frac{r + s}{q + r + s}`

likewise, we can calculate the similarity (asymmetric binary similarity) -

` sim(i, j) = 1 - d(i, j) `

which leave us with

` JC = sim(i, j) = \frac{q}{q + r + s} `

The coefficient sim(i, j) is also known as Jaccard coefficient.

Graphical Display of Basic Stats of Data

by Unknown / in Analysis, data, Data Mining, Display, Graphical, Statistics, Visual, Visualization /

After a long time, got a chance to share somethings with you guys, so feeling awesome :-), Today we are gonna see the Graphical Display of Data Stats or sometime we call it Exploratory Data Analysis as well, This is the best way to understand your data in very less time and set your analysis path for it. So without doing more chats, let's start -

1. Scatter Plot

Very Basic, Very Easy and Most Used EDA(Exploratory Data Analysis) technique. It is 2-D plot between X and Y variables where X or Y can be numeric data features or columns.
With this plot we can easily see if there is any relationship, pattern or trends between between these 2 features or any data outlier existing. It is also useful to explore possibility of correlation relationships. Correlation can be positive, negative or neutral.

Now, let's look into a scatter plot -
I am using IRIS dataset and Python matplotlib library for this illustration - -

2. Histogram

Histogram plot is one of the oldest plotting technique to summarize the data distribution of a attribute X. X can be numerical feature and height of bar is frequency. Resulting plot is also called Bar Chart.

3. Quantile Plot (Bar Charts)

Quantile Plot or Bar Charts also used to display the uni-variate variables data distribution as well as plot the percentile information with outlier detection.

Keep looking for this space for further update.

Happy Learning

Check if Python Pandas DataFrame Column is having NaN or NULL

by Unknown / in Analysis, Analytics, Cleanse, data, Data Mining, dataframe, Exploration, IPython, Jupyter, Python /

Before implementing any algorithm on the given data, It is a best practice to explore it first so that you can get an idea about the data. Today, we will learn how to check for missing/Nan/NULL values in data.

1. Reading the data
Reading the csv data into storing it into a pandas dataframe.

2. Exploring data
Checking out the data, how it looks by using head command which fetch me some top rows from dataframe.

3. Checking NULLs
Pandas is proving two methods to check NULLs - isnull() and notnull()
These two returns TRUE and FALSE respectively if the value is NULL. So let's check what it will return for our data

isnull() test

notnull() test

Check 0th row, LoanAmount Column - In isnull() test it is TRUE and in notnull() test it is FALSE. It mean, this row/column is holding null.

But we will not prefer this way for large dataset, as this will return TRUE/FALSE matrix for each data point, instead we would interested to know the counts or a simple check if dataset is holding NULL or not.

Use any()
Python also provide any() method which returns TRUE if there is at least single data point which is true for checked condition.

Use all()

Returns TRUE if all the data points follow the condition.

Now, as we know that there are some nulls/NaN values in our data frame, let's check those out -

data.isnull().sum() - this will return the count of NULLs/NaN values in each column.

If you want to get total no of NaN values, need to take sum once again -

data.isnull().sum().sum()

If you want to get any particular column's NaN calculations -

Here, I have attached the complete Jupyter Notebook for you -

If you want to download the data, You can get it from HERE.

The Three M in Statis : Measures of Center

by Unknown / in Analysis, Analytics, Concept, Data Mining, Statistics, Stats /

In Statistics, 3M summary is very important as it tells a lot about data distribution. These Ms are - Mean, Median and Mode

Mean - Average
Median - Middile Value
Mode - Frequent Item count

You can look into "SUMMARY STATISTICS IN DATA ANALYSIS"
for the calulations.

Summary Statistics in Data Analysis

by Unknown / in Analysis, Analytics, data, Data Mining, Mining, Statistics, Stats, Summary /

Summary statistics are numbers that summarize properties of the data. i.e - Mean, Spread, tendency etc. We will see each one by one.

Let's take a input dataset -

Input: 45, 67, 23, 12, 9, 43, 12, 17, 91
Sorted: 9, 12, 12, 17, 23, 43, 45, 67, 91

Frequency: The frequency of an attribute value is the percentage of time the value occurs in the data set.
In our dataset, Frequency of 12 is 2.

Mode: The mode of a an attribute is the most frequent attribute value

Mode for our dataset is 2 as 12 is the most frequent item which occurs 2 time

Things to remember:
i- There is no mode if all the values are same
ii - Same is applicable if all values occurrence is 1

Usually, Mode and Frequency are used for categorical data

Percentiles: This used for continuous data.
Given an ordinal or continuous attribute x and a number p between 0 and 100, the pth percentile
is a value x_p of x such that p% of the observed values of x are less than x_p.

How to calculate the Percentile:
1. Count the total item in dataset = N
2. Multiply the percentile p with total no of items = N*p
3. This will give you a no which can be a float or integer
4. If it is a float, round off it to nearest integer, named p^th no
i. Sort the data into increasing order
ii. Now, p^th no in this dataset is your percentile value
5. If it is an integer no
i. Sort the data into increasing order
ii. Now, average of p^th no and (p+1)^th no in this dataset is your percentile value

So when we say, 20% means -
No of items in dataset = 9
No of items which should be less than x_p. - 9*20% = 1.8
Round off this to nearest integer - 2
Our dataset is already sorted in increasing order, so check the 2nd value - 12

likewise, 25%, 50% and 75% is - 9*25%, 9*50%, 9*75% = 2.25^th, 4.5^th, 6.75^th
2^th, 5^th, 7^th - 12, 23, 45

This is one way to calculate the percentile, If you use calculator or some other method, it might be slightly different.

Mean or Average: Sum(all items) / Total no of element

Mean - (9+12+12+17+23+43+45+67+91)/9 = 34.4

However, the mean is very sensitive to outliers. So to understand the data tendency, we go for median rather than means.

Median: Median is 50 percentile, or middle value

How to get Median/Middle value - a. Sort the data into increasing orderb. Get total no of elements - N if N is even - median = ( N/2th element + [N/2 + 1]th element) / 2 if N is odd - median = ceil(N/2)th element

For our case, N = 9, which is odd, so ceil(9/2) = ceil(4.5) = 5th element

Median = 23

Range: Difference between Max and Min is called range.

Input dataset range - 91-9 = 82

Variance: The variance or standard deviation is the most common measure of the spread of a set of points.

`variance(x) = \sigma^2 = \frac{1}{n-1}\Sigma_{i=1}^n(x_i-\bar{x})^2`
where `\bar{x}` is Mean of all value of x
m = total no of items in dataset
`\sigma` is standard deviation

Learning Matplotlib #2

by Unknown / in Analysis, Analytics, IPython, Jupyter, Machine Learning, Matplotlib, Plotting, Python, Tutorial, Visualization /

Learning Numpy #2

by Unknown / in Analysis, Analytics, Code, Computation, data, github, IPython, Jupyter, Learning, Library, Machine, Machine Learning, ML, Notebook, Numerical, NumPy, Pandas, Python, scikit-learn, Tutorial /

Continuing the first post NUMPY #1 .......

Click Here for Next Tutorial ~

Like the below page to get update
https://www.facebook.com/datastage4you
https://twitter.com/datagenx
https://plus.google.com/+AtulSingh0/posts
https://datagenx.slack.com/messages/datascience/

Learning Numpy #1

Numpy is a python library used for numerical calculations and this is better performant than pure python. In this notebook, I have shared some basics of Numpy and will share more in next few posts. I hope you find these useful.

My Learning Path for Machine Learning

by Unknown / in Analysis, Analytics, Code, data, github, IPython, Learning, Library, Machine, Machine Learning, ML, Pandas, Python, scikit-learn, Tutorial /

I am a Python Lover guy so my way includes lots of Python points. If you dont know the basics of this wonderful language, start it from HERE else you can follow the links which I am going to share.

Learning ML is not only studying ML algorithms, it includes Basic Algebra, Statistics, Algorithms, Programming and lot more. But no need to afraid as such :-) we need to start from somewhere.....

This is my github repo, you can fork it and follow me with these 2 links --

Fork - Fork
Follow - Follow @atulsingh0
I am still updating this list and welcome you to update this as well.