{"id":1316,"date":"2019-02-15T10:32:04","date_gmt":"2019-02-15T10:32:04","guid":{"rendered":"http:\/\/kusuaks7\/?p=921"},"modified":"2023-07-31T08:56:09","modified_gmt":"2023-07-31T08:56:09","slug":"data-analytics-with-python-by-web-scraping-illustration-with-cia-world-factbook","status":"publish","type":"post","link":"https:\/\/www.experfy.com\/blog\/bigdata-cloud\/data-analytics-with-python-by-web-scraping-illustration-with-cia-world-factbook\/","title":{"rendered":"Data Analytics with Python by Web scraping: Illustration with CIA World Factbook"},"content":{"rendered":"

Ready to learn Data Science? Browse courses<\/a>\u00a0like\u00a0Data Science Training and Certification<\/a> developed by industry thought leaders and Experfy in Harvard Innovation Lab.<\/em><\/strong><\/p>\n

In this article, we show how to use Python libraries and HTML parsing to extract useful information from a website and answer some important analytics questions afterwards.<\/h4>\n

\"experfy-blog\"<\/p>\n

In<\/strong>a data science project, almost always the most time consuming and messy part is the data gathering and cleaning. Everyone likes to build a cool deep neural network (or XGboost) model or two and show off one\u2019s skills with cool 3D interactive plots. But the models need raw data to start with and they don\u2019t come easy and clean.<\/p>\n

Life, after all, is not Kaggle where a zip file full of data is waiting for you to be unpacked and modeled\u00a0\ud83d\ude42<\/em><\/strong><\/p>\n

But why gather data or build model anyway<\/strong>? The fundamental motivation is to answer a business or scientific or social question.\u00a0Is there a trend<\/em>?\u00a0Is this thing related to that<\/em>?\u00a0Can the measurement of this entity predict the outcome for that phenomena<\/em>? It is because answering this question will validate a hypothesis you have as a scientist\/practitioner of the field. You are just using data (as opposed to test tubes like a chemist or magnets like a physicist) to test your hypothesis and prove\/disprove it scientifically.\u00a0That is the \u2018science\u2019 part of the data science. Nothing more, nothing less\u2026<\/strong><\/p>\n

Trust me, it is not that hard to come up with a good quality question which requires a bit of application of data science techniques to answer. Each such question then becomes a small little project of your which you can code up and showcase on a open-source platform like Github to show to your friends. Even if you are not a data scientist by profession, nobody can stop you writing cool program to answer a good data question. That showcases you as a person who is comfortable around data and one who can tell a story with data.<\/p>\n

Let\u2019s tackle one such question today\u2026<\/p>\n

Is there any relationship between the GDP (in terms of purchasing power parity) of a country and the percentage of its Internet users? And is this trend similar for low-income\/middle-income\/high-income countries?<\/em><\/strong><\/p>\n

Now, there can be any number of sources you can think of to gather data for answering this question. I found that an website from CIA (Yes, the \u2018AGENCY\u2019), which hosts basic factual information about all countries around the world, is a good place to scrape the data from.<\/p>\n

So, we will use following Python modules to build our database and visualizations,<\/p>\n