Skip to content

Commit b07a0d6

Browse files
committed
Update COBOL Programming Course #2 - Advanced Topics.md
Signed-off-by: Radha Venkatesan <[email protected]> Signed-off-by: radha <[email protected]>
1 parent 099ccd6 commit b07a0d6

File tree

1 file changed

+2
-2
lines changed

1 file changed

+2
-2
lines changed

COBOL Programming Course #2 - Advanced Topics/COBOL Programming Course #2 - Advanced Topics.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -220,9 +220,9 @@ The site offers a dynamic list of posts/stories, submitted by users, each of whi
220220
We will be working on a Hacker News 2015-2016 dataset from Kaggle with a full year’s worth of stories: Our goal is to extract only the Mainframe/COBOL related stories and assign ranking scores to them based on (a simplified version) the published Hacker News ranking algorithm. We will create a front page report that reflects this ranking order. The algorithm works in a way that nothing stays on the front page for too long, so a story’s score will eventually drop to zero over time (the gravity effect). Since our posts are spread out over a year and as older posts will always have a lower (or zero) ranking, we will distort the data so all our stories have the same date and and consider only the times in the ranking score calculation. This will give all our posts a fair chance of landing the front page. Our front page report is published at 11:59pm. [Here's some additional information on the ranking.](http://www.righto.com/2013/11/how-hacker-news-ranking-really-works.html)
221221
222222
### The Plan
223-
[] There are different creative ways of accomplishing this but here’s our plan: We will have a COBOL program that reads the input CSV file and retrieves only the ***Mainframe/COBOL*** stories. It then calculates the ranking score for the stories by factoring in the time they were posted and the number of votes they received. Each of the records is then written to an output dataset along with the ranking score.
223+
- There are different creative ways of accomplishing this but here’s our plan: We will have a COBOL program that reads the input CSV file and retrieves only the ***Mainframe/COBOL*** stories. It then calculates the ranking score for the stories by factoring in the time they were posted and the number of votes they received. Each of the records is then written to an output dataset along with the ranking score.
224224
225-
[] We will then use `DFSORT` to sort the output dataset on ranking score, highest to lowest and display the posts as a simple report mimicking the front page.
225+
- We will then use `DFSORT` to sort the output dataset on ranking score, highest to lowest and display the posts as a simple report mimicking the front page.
226226
227227
Let's get started!
228228
1. Take a look and familiarize yourself with the dataset on z/OS: `ZOS.PUBLIC.HACKER.NEWS`. This is a CSV file that serves as input to your COBOL program. The file was created by downloading [this Kaggle dataset](https://www.kaggle.com/hacker-news/hacker-news-posts), removing the lengthy `URL` column that is of no relevance to us and uploading it to z/OS. You can directly reference this DS in your JCL. Please avoid making a copy as it is fairly large with around 300,000 records.

0 commit comments

Comments
 (0)