Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
Describe the solution you'd like
A clear and concise description of what you want to happen.
Right now we just display some recently added posts but I want our main pages to display posts based on past posts that users have looked at.
Lets say that an user really likes to look at Sova recon lineups for ascent posts. When they visit the valorant page next time their feed should be filled with sova lineups on ascent. For CS we will just track nade and map.
This means that we should track user activity and process information on their profile in some way. We don't have the capability to do any type of ML right now (maybe in the future) so this is an algorithm that I will propose
1) Make a new entry that tracks the most recent 30 posts that an user has clicked on for each game.
2) Find the most commonly visited agent, map, utility across the 30 posts. With this we can roughly get an idea of what an user is searching for
note if there's not enough posts, we need to default to the searching for the most popular agent and map
note2 for CS we just focus on nade type and map.
3) Fill their feed with those "recommended posts" but default to most recent posts if we run out.
Obviously, this is a very elementary-style algorithm and we will need ML to make it more sophisticated
We will also have to modify the user entry in our DB
Is your feature request related to a problem? Please describe. A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
Describe the solution you'd like A clear and concise description of what you want to happen.
Right now we just display some recently added posts but I want our main pages to display posts based on past posts that users have looked at.
Lets say that an user really likes to look at Sova recon lineups for ascent posts. When they visit the valorant page next time their feed should be filled with sova lineups on ascent. For CS we will just track nade and map.
This means that we should track user activity and process information on their profile in some way. We don't have the capability to do any type of ML right now (maybe in the future) so this is an algorithm that I will propose
1) Make a new entry that tracks the most recent 30 posts that an user has clicked on for each game. 2) Find the most commonly visited agent, map, utility across the 30 posts. With this we can roughly get an idea of what an user is searching for note if there's not enough posts, we need to default to the searching for the most popular agent and map note2 for CS we just focus on nade type and map. 3) Fill their feed with those "recommended posts" but default to most recent posts if we run out.
Obviously, this is a very elementary-style algorithm and we will need ML to make it more sophisticated We will also have to modify the user entry in our DB