Sachin Shrestha

Crime in NSW - How safe is your suburb?

Thu, 17 Nov 2016 00:00:00 +0000

Author: Sachin Shrestha
Data Source: NSW Bureau of Crime Statistics and Research
Image source: www.nsw.gov.au

How safe is your suburb?

Find the trend of crimes in the suburb you live in. Has it increased over the years? Or is it getting safer?

How many home break-ins, robberies, motor vehicle thefts, store thefts or drug-related offences were reported in your suburb in the years 2011 - 2015?

How does the monthly crime numbers look like? Is there a pattern? In what months do crimes happen the most? When do they occur the least?

Your answers are in the following dashbboard.

How do people travel to work in New South Wales?

Fri, 08 Jan 2016 00:00:00 +0000

Market Segmentation for Frequent Flyer Program

Wed, 25 Nov 2015 00:00:00 +0000

Market segmentation¹ is a marketing strategy which involves dividing a broad target market into subsets of consumers, businesses, or countries that have, or are perceived to have, common needs, interests, and priorities, and then designing and implementing strategies to target them. Market segmentation strategies are generally used to identify and further define the target customers, and provide supporting data for marketing plan elements such as positioning to achieve certain marketing plan objectives. Businesses may develop product differentiation strategies, or an undifferentiated approach, involving specific products or product lines depending on the specific demand and attributes of the target segment.

In this project, I will analyse data from an ariline’s frequent flyer program to group it’s customers into different market segments. In particular, I will use clustering algorithm to segment the airline’s market into different clusters.

The data is sourced from www.dataminingbook.com.

The airlines dataframe contains 7 variables all of which are integers. The variables and their description are shown in the following table.

Name of Variable	Variable Type	Description
Balance	Integer	Frequent flyer points earned so far
QualMiles	Integer	Number of qualifying miles
BonusMiles	Integer	Miles earned from non-flight transactions
BonusTrans	Integer	Number of non-flight transactions
FlightMiles	Integer	Miles earned from flight transactions
FlightTrans	Integer	Number of flight transactions
DaysSinceEnroll	Integer	Number of days since joing the frequent flyer program

The details of the modelling process and the R code to this project is avaiable here.

Results from the model shows that the airline’s customers (that are members of its frequent flyer program) can be categorised into five distinct groups. Analysis of each individual group shows that the groups may be broadly described as follows:

Infrequent but loyal customers
Customers with large amount of miles mostly from flight transactions
Customers with large amount of miles mostly from non-flight transactions
New customers accumulating miles from non-flight transactions
New and infrequent customers

Here’s the complete R code to the Market Segmentation for Frequent Flyer Customers Project.

¹Wikipedia

What determines whether someone earns more than $50K?

Mon, 09 Nov 2015 00:00:00 +0000

We keep wondering what factors play a role in a person’s annual income. We may think of hundreds of factors, but are there that many?

Here I use publicly available census data (source: UCI Machine Learning Laboratory) to find what factors actually determine whether someone earns more than 50K annually.

The census dataframe used for this project contains 13 variables - 12 independent variables (features) and one dependent variable (label). The variables are shown in the table below.

The variable fiftyKPlus is the variable to be predicted. So, it is the dependent variable. The remaining 12 variables are used to predict the dependent variable.

Name of Variable	Variable Type	Values
age	continuous	17 - 90 years
workclass	categorical	Federal-gov, Local-gov, Never-worked, Private, Self-emp-inc, Self-emp-not-inc, State-gov, Without-pay
education	categorical	Preschool, 1st-4th, 5th-6th, 7th-8th, 9th, 10th, 11th, 12th, HS-Grad, Assoc-acdm, Assoc-voc, Prof-school, Some-college, Bachelors, Masters, Doctorate
maritalstatus	categorical	Divorced, Married-AF-spouse, Married-civ-spouse, Married-spouse-absent, Never-married, Separated, Widowed
occupation	categorical	Adm-clerical, Armed-Forces, Craft-repair, Exec-managerial, Farming-fishing, Handlers-cleaners, Other-service, Priv-house-serv, Prof-specialty, Protective-serv, Sales, Tech-support, Transport-moving
relationship	categorical	Unmarried, Husband, Wife, Not-in-family, Other-relative
race	categorical	White, Black, Amer-Indian-Eskimo,,Asian-Pac-Islander, Other
sex	categorical	Female, Male
capitalgain	continuous	0 - $100,000
capitalloss	continuous	0 - $4,356
hoursperweek	continuous	1 - 99 hours
nativecountry	categorical	Cambodia, Canada, China, Columbia, Cuba, Dominican-Republic, Ecuador, El-Salvador, England, France, Germany etc.
fiftyKPlus	categorical	Less than or equal to $50,000, Greater than $50,000

The dependent variable fiftyKPlus has two classes: <=50K or >50K. Therefore, this being a classification problem, I have developed the following models to predict whether or not a person earns more than 50K annually. I have used R to build the models.

A CART (Classification and Regression Tree) model
A CART model with cross-validation
A Random Forest model
A Logistic Regression model

The details of the modelling process and the R code to this project is avaiable here.

To decide which model is best suited to predict whether or not a person earns more than 50K a year, the performance of each model has to be evaluated.

Performance of the models:

Type of Model	Baseline Accuracy	Model Accuracy	AUC
CART		0.849582	0.846746
CART with cross validation		0.862403	0.871933
Random Forest		0.825580	NA
Logistic Regression	0.759362	0.850989	0.905733

Since the accuracy of the random forest model is the least of the four models, it may not be best suited for predicting the label in this project.

The logistic regression model does a little better than the CART model. However, the logistic regression model is not easily interpretable. Although the model coefficients may be used to determine the significance of features, the coefficients do not offer simple explanation of how decision is made. For example, look at the following results for the variable education from the summary of the logistic regression model.


occupation Adm-clerical                  -7.306e-02  1.283e-01  -0.569 0.569102
occupation Armed-Forces                  -1.393e+01  9.937e+02  -0.014 0.988812
occupation Craft-repair                   4.995e-02  1.092e-01   0.457 0.647523
occupation Exec-managerial                7.267e-01  1.133e-01   6.414 1.42e-10 ***
occupation Farming-fishing               -1.214e+00  1.851e-01  -6.558 5.46e-11 ***
occupation Handlers-cleaners             -7.398e-01  1.879e-01  -3.938 8.22e-05 ***
occupation Machine-op-inspct             -3.752e-01  1.393e-01  -2.693 0.007089 **
occupation Other-service                 -9.234e-01  1.655e-01  -5.579 2.42e-08 ***
occupation Priv-house-serv               -1.390e+01  2.166e+02  -0.064 0.948837
occupation Prof-specialty                 3.865e-01  1.213e-01   3.186 0.001442 **
occupation Protective-serv                6.512e-01  1.718e-01   3.790 0.000150 ***
occupation Sales                          1.892e-01  1.170e-01   1.617 0.105965
occupation Tech-support                   5.251e-01  1.532e-01   3.428 0.000608 ***

Observe how some sub-categories of the variable have been marked as significant (with three asterisks) whereas some are marked as not significant at all. A similar trend can be seen with other variables too. This is complicated! The model is difficult to use to quickly make a prediciton for a new case. In summary, the logistic regression model is not easily interpretable.

On the other had, the CART model is more easily interpretable. A CART model is also preferable because it does not assume a linear model like a logistic regression model.

As seen in the plot of the CART model (see code), the features that split the tree are relationship, capitalgain and education. The CART model is more interpretable in the sense that the model tells that these three features are the strong predictors of whether or not a person earns more than 50K annually.

The CART model with cross-validation has the highest accuracy of the models. Although the plot of the CART model with cross-validation presents a tree that is more complex than that of the CART model without cross-validation, a closer look shows that both the trees have been split by the same three features: relationship, capitalgain and education.

We may thus conclude that, out of the four models, the CART model with cross-validation best predicts whether or not a person earns more than $50K a year.

And what factors are the strong predictors? Well, as we just saw, the following three factors most signify whether a person earns more than $50K every year:

relationship,
capitalgain and
education.