I am also a Joplin user. I found a Joplin plugin that - I think - might be very nice to have in Obsidian: a plugin to extract keywords from note and assign them as a note’s tags.
See joplin-tags-generator
14 Likes
Might be useful! Especially, if you could specify the tags it creates
1 Like
I would love this very much! I am importing my whole life into Obsidian as we speak, and it would be massively useful to have a plugin that would search through my whole vault, find likely tags (by identifying strings that appear frequently in the vault, for example) – or assign tags to strings I specify. In fact, I’m not entirely sure I can make the most of Obsidian until there’s such a plugin, unless I’m missing something. (And if I am, please tell me!)
2 Likes
Others have repeatedly made a feature request for automated concept tagging in notes (see the end of this post).
As of now, there’s no way to automatically tag ideas (whether a given idea is expressed using one or more words) in Obsidian. Instead of adding notes and having them undergo automatic tagging, we users have to invest time into manually tagging every note we create.
As anyone can imagine, this is highly time-consuming --especially if you have a big vault, of let’s say 5GB. And it’s not clear or it’s doubtful whether there is any value to engaging in manual tagging.
What’s worse, adding the hashtags needed to create tags changes the way that notes can be searched, in making the process more complex or less accurate. For instance, searching tags and words require different inputs into search boxes.
Is there a way to automate the tagging process in Obsidian, whether using NLP tools or AI?
Startups like Flowdex are building ecosystems based primarily on making the tagging process easier. Its claim is, “Flowdex automatically tags your notes using AI so you can find what you need in a few clicks.”
Flowdex advancing this value proposition shows there’s a need for automated tagging. Some users have said they wouldn’t use Flowdex because of data storage issues (as discussed here: https://www.reddit.com/r/DigitalGardens/comments/w5c8fz/introducing_flowdex_a_new_way_to_organize_your/).
While Obsidian’s local storage and other plugin-enabled features are valuable, a different ecosystem can overshadow Obsidian, such as with automated tagging and much smarter search capacities (as exemplified by Google).
Will the Obsidian community respond by making tagging easier/smarter, either through a feature addition or plugin?
1 Like
It should be noted that, so far, all requests for automated tagging have been ignored by Obsidian and the Obsidian community.
I’m not sure what the point of automatic tagging is. To me tagging has to be manual to make any sense.
Instead it might just make more sense to improve on searching rather than preemptively adding structure in an imperfect way.
2 Likes
This Flowdex YouTube video shows some of the benefits: Flowdex Demo - YouTube
And though you may not see the point, there are many people who do. Manual tagging presents some major issues around time consumption and search.
Lastly, like Obsidian, Flowdex also uses a markdown editor.
I’m not sure which is best --the Logseq plugin you just referenced, or the one for Joplin (which has been discussed and mentioned by another user in this post: Automatic Tag Generation).
The Joplin plugin, which generates tags, can be found here precisely: GitHub - forcewake/joplin-tags-generator: Plugin for Joplin which can be used to extract keywords from note and assign them as a note's tags.
The fact that Joplin, Logseq, and other emerging apps offer this capacity shows there’s a need.
How do we successfully put forward a feature request as a community?
This might be implemented sooner if you move it to Plugin Ideas. After all, that’s how it became available in Joplin and Logseq.
It was in plugin idea. I am gonna move it back there.
If more than 20 people want a plugin like this, I can develop one.
Reply to this message if you want.
12 Likes
To be fair, auto tagging could start simpler. Having the ability to mark a note as MOC with a given tag (say #physics), and then having all notes linked to it created with the same #physics tag would already streamline quite a bit of the process.
I would absolutely love this plugin, I have so many research articles that connect with each other it would be amazing to automatically tag them
I copy paste the note in chatGPT and ask it to generate tags which I manually add. Having that automated would be a great start.
This plugin : GitHub - louis030195/obsidian-ava: Quickly format your notes with ChatGPT in Obsidian includes a “Suggest Tags” command.
1 Like
I found it !! “Auto Classifier” Plugin in community plugin is what you are saying.
It is an Obsidian plugin that helps you automatically classify tag from your notes using the ChatGPT API. The plugin can analyze your note (It can be title, frontmatter, content or selected area) and suggest relevant tag based on the input with tags in your valut.
This can be used for various and specific purposes, for example, DDC classification for books, keyword recommendation, research paper categorization and so on. Save you time and improving your note organization.
You can download at community plugin, or Git-Hub(GitHub - HyeonseoNam/auto-classifier: Auto classification plugin for Obsidian using ChatGPT.)
1 Like
This is really cool, but I use Obsidian for note-taking at work. I can’t submit my letters to chat gpt. It would have to be local. I would like to even do something more simple. Maybe a Tag ID file where I set a tag and apply a custom keyword Definer. This could also be used as ref links to other folders or files. So if I have a file named DevIdea1 and in another file, I say DevIdea1, it just auto links the 2. (Maybe that last one is already a thing, I don’t know.)
When taking notes, I often forget to go back and manually add these connections. It would be nice if they “Just happened.”
1 Like
I managed to modify the plugin that uses OpenAI for tags (GitHub - HyeonseoNam/auto-classifier: Auto classification plugin for Obsidian using ChatGPT.) and instead of OpenAI it now calculates the most frequent words and uses them as tags. Just keep it on “Demo” mode, and instead of using the random demo tags, it will calculate the most frequent words.
HOW TO INSTALL:
Just install the plugin above as usual, and don’t use any OpenAI API Key. Open the plugin folder and edit main.js code in the following way:
Add this code above “function getPrefix(input, prefixCharacters = DEFAULT_PREFIX_CHARACTERS) {”:
And add this code between “if (settings.demoMode) {” and “} else if (settings.openaiApiKey.length > 0) {”:
Then just restart Obsidian and it should work.
Maybe someone can put it on Github so it is easier to use for non-coders. Enjoy!