Thirukkural-Dataset

[LavanyaMaruthachalam]

Hello sir. Iam Lavanya Maruthachalam studying MSc Data science in University of Roehampton.
Currently I'm doing my final project which is creating chatbot for Thirukkural and Athichudi.
I came across your GitHub profile for the creating dataset which is absolutely awesome. The structure of the dataset is very organized and that's how I wanted to create my own dataset.
I'm trying to combine all the explanations so far given to Thirukkural and Athichudi. But failing in my every attempt. I extracted the 5 different explanations for Thirukkural from PDF files, but I couldn't clean and make it as a structure.
I know manually arranging and structuring is impossible, but I want to know how you achieved it sir.

[Purus]

Hope this is useful for you. You can take the json from this repo and use the jq tool to extract and restructure as per your need. . Just a quick thought.

[LavanyaMaruthachalam]

Definitely it is helpful Sir. Thank you for your reply

[vijayanandrp]

@LavanyaMaruthachalam - Read the json file from data folder using python. Explanation is in the key - value. If you know how to handle dictionary its easy to extract only explanations.

[LavanyaMaruthachalam]

Thank you so much for your approval and path to use your data. Just one more question sir, may I know how did created the text file at starting for every chapters.
I extract the text from Project Madurai using OCR method but I am failing in break it down and sorting it into chapters. May I know how you created the whole dictionary is it from scratch.

[vijayanandrp]

I have written a python web scrapper to extract the data from sites.

[LavanyaMaruthachalam]

Oh, Okay sir, Thank you so much for your support, sir.