rishab-sharma / youtube-data-extraction-and-analysis

Collect data using Youtube API v3 and analyse using Google Charts.

PreCog-Task

1. Data extraction using the search_keyword.py script:

Using the Youtube API v3, data has been extracted for a particular keyword by giving command-line parameters. I have collected data for ~500 videos for a particular keyword (in my case "ted").

The data extracted includes video's "title","description","videoId","viewCount","likeCount","dislikeCount","commentCount" and channel's "channelTitle","viewCount","commentCount","subscriberCount". As youtube allows only 25 results/page, I have used pagination using the "nextPageToken" to get the next page results.

To use the script, in terminal run the following command: $ python search_keyword.py --q keyword_to_search

2. Webapp for the data analysis:

Data collected in part 1 has been analysed using the Google Charts. To check the webapp, open it on a browser or host it on local server using the following command in the directory where the analysis.html is located: $ python -m SimpleHTTPServer. This would host the webapp on the local server on the address "localhost:8000". Page is also hosted on: https://youtube-analysis-precogtask.herokuapp.com/.

3. Predicting the number of likes on a YouTube video URL one can input:

This task has to be done for the collected data.