Introduction to Computation for the Social Sciences Assignment 7
Prof. Dr. Karsten Donnay, Stefan Scholz Winter Term 2018 / 2019
Please solve the exercises below and submit your solutions to our GitHub Classroom until Dec, 18th midnight. Submit all your code in one executable file (py
/ ipynb ) and your list in one delimited text file (csv).
This assignment sheet is graded according to the following scheme: 3 points (fully complete and correct), 2 points (mostly ), 1 point (partly ) and 0 points (hardly). You will get individual feedback in your repository.
Exercise 1: Simple Web Crawler
Web Crawling, i.e. the automated access of Web resources via software, and Web scraping, i.e. the extraction of content from Web resources, are essential techniques to gather data for many analysis tasks in the social sciences. We will implement a very simple Web crawler that starts at a given URL and iteratively accesses all links, more specifically all href-tags in the HTML source code of a Website.
To get an idea how to approach this problem using Python, have a look at the sample implementation of a crawler in the subchapters Traversing a Single Domain and Crawling an Entire Site in the book Web Scraping with Python by Ryan Mitchell. If you search for the book online, you will find a PDF version of it.
Implement an iterative crawler (no recursive calls) that:
-
opens the front page of the German Wikipedia1 and downloads the html resource
-
parses the html file as a BeautifulSoup object.
-
collects all internal href tags, i.e. only links to other Wikipedia pages on the server de.wikipedia.org
-
opens each of the collected links and parses the returned html resource for additional internal links
Additionally, your crawler has to meet the following requirements:
-
Consider the Wikipedia front page as the root (level 0) of a tree and the Wikipedia pages linked to on the front page as level 1. Crawl no deeper than level 2!
-
Visit each unique link only once, i.e. disregard links that you traversed before
-
Store all links of the same level in a list, i.e. one list per level.
-
Include links that you encounter more than once in the list of the lowest level (closer to the root) they appear at.
-
Write the lists of links to a CSV file in the format level ; link_URL.
Finally, submit your solutions by adding your code and csv file in your Git repository within the folder assignment07 > solution. Do not forget to commit and push your solutions afterwards.