Looking for advice on how to approach a feature

Im making a recipe database website where the idea is to group together similar recipes for the same dish and the most popular version of the dish is what is return on search results. Ive got a system in place that assigns a score to a recipe based on user's interactions with the recipe (saving, commenting, rating, etc). Also using the ancestry gem to create a heirarchy of the recipes based on that score. What I'm having issues trying to figure out is a way to initially group the recipes into a heirarchy when they are created. I've got a couple thoughts on approaches but nothing figured out yet. Here are my main two thoughts:

Looking mainly for a discussion on ways to approach this problem. The ultimate goal is to create a recipe site that shows you the most popular version of a recipe, while allowing you to still make changes or look at other options.

You are not going to like my idea, but I think the best way to handle this is to do it manually. Each night, just log in and do housekeeping. Once you have a few hundred recipes, the number of new one should go down drastically and people are simply going to create variation of existing ones. 

The amount of time you are going to spend trying to build a system to do it automatically is probably not worth the effort. You better invest this time in polishing the rest of the app. 

As developers, we see code as the solution to everything and it makes us lose sight of the end goal.

Yeah Ive been leaning towards changing how I manage this and the tweaking the concept some.  The ultimate goal is to only show the highest rated version of the recipe and promote community changes to it.  So tweaking the interaction some to something I can manage

Alain Pilon - nice angle!  I always get my thoughts down and then try to 'back up' and take a global look at the app and the end-goals, to make sure I am not going overboard with my solutions(which I go overboard about 100% of the time) in order to achieve the goal.  I have found myself trying to figure out how to implement AI on something that never should have had AI to begin with.  Weeks later I had to ask myself, 'what the hell are you doing?'.

I now use the KISS methodology, least amount of moving parts to get something to initial production release and achieve the singular main goal, then start adding the features that were 'glorious' that you thought about during the build process.

Brandon B. I am doing the same thing. I am reducing the number of feature I am implementing to the bare minimum of what will be actually useful then spend about as much time to clean up the UI and make sure to look at the app from a newb point of view. 

If you asked me last week when I thought my app would be release, I would have said 2 weeks. If you ask me today, I will say 2 weeks again ;-) At least now I am wasting time trying to find problem pictures, icons and colors. And an even bigger time sink: writing the landing page, about, etc... 

All that being said, cant wait for it to be in a presentable state to show it off ;-)

Yeah I agree, sometimes my desire to make a completely hands off website that will allow me to just sit back and tweak features and the look goes a little crazy....I guess it wouldnt be too hard (at least at the start when there arent many users) for me to manually go in and look at the recipes and assign it to a parent recipe (if one matches).

And I feel you with the landing page, about, etc.... I hate doing those so much