25 karma · joined August 17, 2022
You can check out my blog, it will answer all your questions -
https://medium.com/@19aditiyadav/how-i-built-bookclub-ai-ai-...
I have build an AI book-recommendation website this month & have 30k books in my database right now.
Bookclub.ai has based on genres, celebrity recommendations, popular lists and series.
I have applied AI to present a visual representation of "Books More like this" in the form of mind-map feature.
One of the best features is "Book-Genie" which recommends books real time based on your mood.
I am a sole member of this project right now and in bootstrap phase. Would love for you to checkout my website and any feedbacks/collaborations are welcomed.
Happy reading!
Link - bookclub.ai
My Email id (for feedbacks/collaborations) - 19aditiyadav@gmail.com
Are you tired of spending hours scrolling through endless book lists, only to end up with a stack of books that don’t quite fit your taste? Look no further! Our Book Genie feature is here to help you find the perfect book for you, with just a simple natural language query.
Bookclub.ai revolutionized book industry by launching Book-Genie where users can ask it anything and would get most relevant books. Gloria, a regular Bookclub user and an avid book reader commented “I just love this feature, earlier I had to go specifically for a book recommendation or find keywords from other reviews and do those similarity calculations in my head. Now I can just search by typing “Life in France during Nazi Occupation” and BookGenie returns highly relevant books like The Nigtingale, Book Thief etc”. Matt, a Greek origin immigrant had high praises for this feature where he said “I love reading Greek books and searching in my mother tongue but none of the other websites could understand my intent, Book genie not just finds relevant books but also suggests other lists which are relevant to my query. I used to visit Reddit a lot and sometimes had to wait few hours even for simple suggestions, however I can just play around in real time now”. This feature has resonated well with our beta testers where they see live recommendation refresh based on their upvotes of surfaced books. Brandon, who associates book reading with community is excited to see what others are searching in real-time and quality of their recommendations.
Book Genie has been trained on over 5 billion parameters, using the state-of-the-art BERT (Bidirectional Encoder Representations from Transformers) model. This means that it has the ability to understand and interpret human-level queries, and provide recommendations based on your unique interests and preferences.
So how does it work? It’s simple! Just type in a query like “I’m in the mood for a mystery novel set in the 1920s” or “Can you recommend a non-fiction book about the history of artificial intelligence?”, and Book Genie will use its advanced natural language processing capabilities to understand your request and provide a list of tailored recommendations.
It is different from its competitors like Amazon or Goodreads in that 1.It has an interface tailored for book-discovery, 2.has dedicated category for celebrity recommendations, book series and popular lists, 3.trained on 100k+ Reddit datasets to understand human-level queries and has sub-categories tailored to reflect broad genres. Also the book-detail page breaks down recommendations to showcase multiple book-associations e.g a book Intelligent Investor has multiple recommendation bucket as “Other Investing Books”, ”Other Personal Finance Books” etc.
Book recommendations for Intelligent Investor Aditi Yadav, founder of Bookclub.ai is seeing it as an opportunity to revamp reading experience by bringing State Of the Art research in NLP and machine learning at the fingertips. She considers it as a first step in suite of high impact features planned in 2023 around knowledge graph, computer vision and language based personalization.
So next time you’re looking for your next great read, give Book Genie a try. It’s the perfect personal assistant for book lovers everywhere. Happy reading!
It uses Transformer and BERT models to understands human level queries, maps its into vector space, find the dot product of story-line embeddings and user queries to generate most relevant books in seconds.
Ask something like ‘Women Lead in World War from Paris’ and it suggests titles like ‘Nightingale’.
It has features like Book-Genie which has used StateOfTheArt algorithms to make sense of human queries from 1Mn+ reddit postings for book suggestions (I recount my waiting times of ~1 Month to get data dumps and training models on top of existing NLP models trained for Question and Answers). Mind-Map feature is another NLP product to provide visualizations of MoreLikeThis books in vector space by understanding story line and good-reads reviews and voting data.
Every book has its dedicated detail page where you get a detailed summary of its storyline, the relevant info about author, ratings, review count and most importantly “More-Like-This” books based on genres, subjects, theme etc. It identifies the lists that can be associated with a given books and performs a ranking using ML based embeddings.
Bookclub.ai is inspired by Goodreads and I see it standing in the same space someday where book-lovers from all over the world can connect and hangout, explore books and share their experiences. Bookclub.ai wants to become the de-facto platform for book lovers, designed to help you find the next great book based on what you have previously read.
Hello everyone,
I have build an AI book recommendation website. It 30k books in my database right now.
Books are arranged based on genres, celebrity recommendations, popular lists and series.
I have applied AI to present a visual representation of "Books More like this" in the form of mind-map feature.
One of the best features is "Book-Genie" which recommends books real time based on your mood.
I am a sole member of this project right now and in bootstrap phase. Would love for you to checkout my website and any feedbacks/collaborations are welcomed.
Happy reading!
Link - bookclub.ai
My Email id (for feedbacks/collaborations) - 19aditiyadav@gmail.com
Will be looking forward to feedbacks and suggested features.