Very often I have difficulties finding books to read that are similar to the books that I loved. I'm thinking about creating a project, somewhat like spotify for books but before engaging I want to see whether I'm delusional. Maybe there isn't a real demand for something like that and my perception of reality is only mine. Hope someone sees this and replies, I'd be appreciative
by Front_Push_6466
12 Comments
Totally agree, the problem for me is how the algorithms work on platforms like Goodreads. It’s collaborative filtering aka looks at what you’ve rated highly and then recs books that were also rated highly by other people with similar reading habits.
What happens? Stuck in a cycle of the same. The algorithm focuses so much on ratings and tags, it stays recommending other popular books with same tags.
I’ve had success so far with StoryGraph. Its algorithm focuses on things like mood to give more specific recommendations.
Also, when it doubt, I plug in what I’m looking for into AI and ask for 10 recs and explain why they fit my prompt.
I watch an obnoxious amount of bookish content on YouTube.
Book group I’m in. People can suggest books.
I’ve found [this](https://lithub.com/) to be the best for my tastes.
I follow a couple people on Goodreads who read A LOT and whose taste is in line with mine. It took some digging, but now really has cut down on my “what to read next?” time.
my public library – booklists, displays, and their personalized recommendation service.
StoryGraph and the Libby app (I listen to a lot of audiobooks but it’s almost all nonfiction, so take that with a grain of salt)
Actually TikTok BUT! (big but) it does take a lot of curating of your fyp to stop getting the “popular booktok books” and to get it more in line with your interests. What I usually do is skip a video as soon as I know it’s about a genre I don’t like and try to interact as much as possible with people who do read according to my taste. Youtube as well 🙂
I have never needed a book recommendation in my entire life. I see a lot of folks trying to use AI to create this but they forget that readers generally don’t always want a book that is the same as what they just read. Everyone I know has a TBR list of at least 20 books.
I get my book recommendations from TV, movies, documentaries and references in papers and books.
One of the first books I bought this way was The People’s History of United States by Howard Zinn because Matt Damon mentioned it in one line in the movie Good Will Hunting.
In an interview, Matt said he wanted to include that book in the movie because he was living next door to Zinn when he was a kid, so he was one of the first reviewers of the book. He wanted to tell the world about it And he did. It is a great book by a great man.
Reddit groups! Have discovered books I wouldn’t have normally chosen from various user suggestions.
I’m curious how a Spotify for books would be different from Goodreads. I would be very interested, but having a hard time envisioning it and would genuinely love to hear more about the ideas floating in your head.
Until then, Goodreads does ok for me. I don’t pay much attention to what the app itself suggests, but I look at new releases and most read this week in my different favourite genres. I still very much judge by cover, and from there will read the description, put it on hold at my library, and then read a couple pages when it comes in. Same as if I were browsing in a library or bookstore. And I have found a few people to follow whose tastes are similar to mine and get introduced to new books via their reviews.
What doesn’t work for me in Goodreads is the crowd sourced rating. I find too many books have 4+ ratings that don’t seem to warrant it, while conversely I’ll shy away from something that might have a 3.5 rating but then if I do end up reading it, feel that its rating should be much higher.
Because of this I have been going to BookMarks – it’s an aggregation of critic reviews similar to Metacritic. Every book that I’ve chosen from BookMarks because it had “rave” reviews has been a winner for me.