Hacker_News_Datapipeline

Summary: A ,simple and DAG based, data pipeline filtering, cleaning, aggregating, summarizing json data from Hacker News. This project is a guided project from Dataquest.io

file_to_json: loads 'hn_stories_2014.json' file into a dictionary and return it

filter_stories: return a generator of filtering popular stories that have more than 50 points, more than 1 comment, and do not begin with 'Ask HN'

json_to_csv: build a virtual csv file and put the transformed json data onto the file

extract_titles: returns of every Hacer News story title

clean_titles : get rid of punctuation and transform each title into lower case and then return it

build_keyword_dictionary: return a dictionary of the word frequency in a title

top_keywords : sort the result from build_keyword_dictionary decreasing order then return first 100 results.