Geospatial Analysis and Social Analytics Assessment

Learning Outcomes

This assessment is designed to demonstrate a student’s completion of the following Learning Outcomes:

• Demonstrate an understanding of concepts underlying geospatial analysis and apply them on real life data.

• Carry out social analytics in combination with geospatial data, applying appropriate techniques on social information.

• Design, prototype and implement geospatial applications.

• Identify and describe emerging technologies and research areas relevant to geospatial analytics. Assessment Requirements / Tasks (include all guidance notes)

1.1 Review on different techniques used for geospatial analysis and social media information analysis

Please perform a detailed review on the following applications,

- QGIS

- Open Street Map

- Twitter

- GeoPandas

You are free to use the available resources (academic papers, articles, official documentation etc.) for your research however the sources must be cited appropriately.

Your review should cover the following aspects of each application,

- Brief introduction

- Functionality

- Common use in different areas (business, social study, public health etc.)

- Strengths and limitations

Report on the importance of geospatial analysis in different areas and how these areas are benefiting from geospatial analysis. Among the prominent areas of applications for geospatial analysis – Sustainable

Development, Public Safety, Health & Human Services, Transportation, and Education

Apply geospatial visualisation tool (e.g. GeoPandas) on the dataset provided

This task requires you to use the two datasets (world population and world GDP) accessed from the World Bank. Both of these datasets are available on the Moodle under the Assessment folder. Use the GeoPandas or similar visualisation tool to plot a set of choropleth maps representing the world GDP per capita for the years 1995, 2005, and 2015 respectively. This task requires you to use both the datasets simultaneously for calculating the GDP per capita.

The solution for this task should describe all the major steps taken for generating the choropleth maps. If a Python based tool like GeoPandas is used then the solution should be in a Jupyter notebook form (.ipynb), wherein all the functions, libraries and coding steps should be explained in a lucid manner. In this case, the major steps for generating the choropleths would typically involve, importing the datasets using appropriate Python libraries, data cleaning, geospatial operations, and plotting. The Jupyter notebook should be able to reproduce the choropleth maps without any error. If some other non-Python based visualisation tools are used, then the solution for this task should include a written description about the major steps undertaken for generating the choropleth maps as well as an appropriate number of supporting screenshots should also be presented.

Assessment Requirements / Tasks (include all guidance notes)

2.2 Analyse the datasets and answer specific questions. For plotting within this section, you can use any visualisation tool.

• For year 2015, plot the GDP per capita for only the countries having population greater than 300000000. Very briefly interpret the generated plot.

• For year 2015, plot the GDP per capita for only the countries having population less than 70000000. Very briefly interpret the generated plot.

• For year 2015, plot the GDP per capita for only the countries having gross GDP between 450000000000 US Dollar and 8920000000000 US Dollar. Very briefly interpret the generated plot.

• What is the percentage change in the GDP per capita from 1995 to 2015, for the country having the highest population in 2015?

• Plot the mean per capita GDP (from 1995 to 2015) of all the countries. Very briefly

 nterpret the generated plot.

• Present a correlation plot between mean population of each country and mean per capita GDP (from 1995 to 2015). Very briefly interpret the generated plot.

NOTE: For the Task 2.2, all the geospatial plots can be presented using choropleth maps or even using simpler heat maps. If a heat map is used for the visualisation then the countries names must clearly be visible on the plots.

The solution for the should be presented in a Jupyter notebook (.ipynb), if a Python based tool is used. All the functions, libraries and coding steps should be explained in a lucid manner. The notebook should run without any error and all the results should easily be reproducible. Your brief interpretation about the generated plots should also be

contained in this Jupyter notebook. If some other non-Python based visualisation tools are used, then the solution for this task should include a written description about the  major steps undertaken for generating the plots as well as an appropriate number of supporting screenshots should also be presented.

Social analytics In this task, you will apply sentiment analysis to Twitter data using the Python libraries TextBlob and Tweepy. Your analysis should cover the following major steps:

• Get 500 tweets on the topic, #Lockdown or #CovidLockdown with a Python script.

• Clean the tweets. Such as, removal of URLs from the tweets.

• Calculate the polarity values of the individual tweets and present them using a suitable visualisation such as, histogram.

• Analyse the public sentiments about the chosen topic (#Lockdown or

#CovidLockdown) based upon the polarity values and make your recommendation about any future lockdown measures based upon the performed analysis.

The solution for the Task 2.3 should include a Jupyter notebook (.ipynb) describing all the major steps performed during the analysis. All the functions, libraries and coding steps should be explained in a lucid manner. The notebook should run without any error and all the results should easily be reproducible. Your interpretation about the results and recommendation for any future lockdown measures should also be contained in this Jupyter notebook.