This is a corpus that we have built using the book reviews from Goodreads. We have scraped the reviews for books along with the genre, the title, rating and the author. In order to build upon this data and create an annotated corpus, we have labelled the specific emotion of a review. These reviews can range from a couple of sentences to a full paragraph with a maximum of 300 words. We have focused only on English reviews. In order to narrow down the reviews that we look at, we chose several distinct genres and, for each genre, we chose book, scrap the reviews and then selected the books recommended by Goodreads in order to find more reviews. This data has internal structure because along with the review, we can extract the book title, the author, genre and the rating given by the user. We have annotated this data using mechanical turks and each review is labelled as one of the following emotions: enthusiastic, sad, bored, disappointed, content, love, neutral. This corpus is of interest because the labelled emotions are distinct from the numbered review and also because it is domain specific because it is text written about books. The results can be used by other people to train their models to recognize certain emotions. It would be interesting to see if certain genres will have reviews with different associated emotions.
Make sure you have Docker.
- To run it locally on your system type the following commands:
- docker pull varadp02/dockerhub:good2
- docker run -p 8000:8000 varadp02/dockerhub:good2
- Go to the browser http://127.0.0.1:8000/ and check if the application is running.
- To shutdown the app, Control + C
- Please read the corpus description carefully.
- Please use the visualisations to examine the corpus statistics.
- Please test the app's search and filter features.