Two days ago, voting started for the BookSmash Challenge, a competition for innovative app ideas in the world of books. The challenge motivated me to realize an idea that I’ve been toying around with for some time now: a book discovery app supported by language technology. ReadTweet takes someone’s Twitter feed, identifies the topics that person tweets about, and then makes book recommendations based on those topics.
Have you ever been at a loss about what to read next? I sure have. With thousands of books published every day, it gets more and more challenging to find the ones you’ll like most. ReadTweet solves this issue with Natural Language Processing. To be precise, it uses a combination of Latent Semantic Analysis and Latent Dirichlet Allocation to match tweets and books on the basis of their semantic profile. For example, it learns that Pope Francis I most often tweets about religion, that Barack Obama tends to talk about society and politics, and that Cristiano Ronaldo is crazy about sports. It then searches its database for books about the same topics, and makes book recommendations.
In the video below, I give a short demonstration of the functionality:
ReadTweet is a proof of concept: it’s neither the most fancy looking nor the most extensive application in the BookSmash challenge, but I’m pretty happy with the result. I realized an old idea of mine, and in doing so, I learnt a bunch about web app development. If you’d like to support ReadTweet in the BookSmash Challenge, you can vote for it here, until September 27th.
Yves Peirsman 