The problem LinkedIn Analyser solves

The project helps a user to do sentiment analysis of someone's profile (let it be Mr. X) in a matter of few minutes. This project will take the username of Mr. X as input and returns a report of how many posts have positive, negative and neutral sentiment value out of all the posts that he had posted. It will also plot a pie chart of the same. The report also consists of individual sentiment analysis of each of the posts. Additionally, the project will also show a scatter plot between the subjectivity and polarity of the posts. After, the sentiment analysis, persona analysis takes place and tags Mr. X's personality into the big 5 personality trait models on the basis of scanned content.

Challenges we ran into

There were a few problems that we ran into like scraping details from an infinite scroll web page on Linkedin which was solved using selenium. Another problem that we faced was during Persona Analysis as none of our team members was familiar with the topic and there was not a lot of data available on the web regarding the correlation between words and the OCEAN traits.
A research paper (Yarkoni T. Personality in 100,000 Words: A large-scale analysis of personality and word use among bloggers. J Res Pers. 2010;44(3):363-373. doi:10.1016/j.jrp.2010.04.001) helped us to understand more about the OCEAN traits and the correlation between words and different traits. There was a liwc analyzer that could have helped us to do the job much efficiently but it was paid. So, to do persona analysis we created a dataset on our own using the knowledge about the topic we gained from the above-mentioned paper.