Can you share more about how the "mood" is determined? Definitely will be controversial and is deserving of a much more clear explanation.
Why am I tense? Is it the sentences I formulate? Upvote vs. downvote patterns? RNG?
I pretty strongly believe that you should take that piece rather seriously, even if you don't previously consider it a big deal. People generally don't like when a stranger tells the world what their mood will be. Especially bad when someone believes it's quite wrong.
Also I love the opportunity for introspection. I'm trying to become a more whole, happy person who isn't defined whatsoever by his career. I would love to learn how some algorithms perceive me and why.
EDIT: Fixed the UI: I'm assuming you were looking at a topic, the "probable mood" used to render as it relates to the topic being searched, not current mood.
Highly recommend reading the "how it works" section (I assume you already have):
That section has a few examples and details. I'll add further details below.
The system ONLY looks at comments sent to it. It, kind of looks like:
{ Author, Comment }
From there, everything else is derived. Nothing else, no profiles, no voting, nothing. That's it.
The system identifies:
- Expertise
- Knowledge
- Interests
- Relevant content
- Promoter score (an improved net promoter score)
- Trends
- Mood
- Related Topics
- etc.
To your question, there are multiple algorithms at work determining mood, most are pretty standard models. These are ones you can find in NLTK, spaCy, CoreNLP, etc. There's also a couple home-baked ones. These models than "vote" for the mood, with an averaged score. Typically these scores are 3 - 7 models voting at any one time. The aggregate score is roughly 80% accurate (based on my tags), and it tags the topics, sentence, and overall comment all separately. Finally, the models are much more accurate at the extremes: Very Positive and Very Negative, but in the middle, it's much fuzzier. This is in part, just how language works, it's subjective.
All NLP models are based on a combination of open datasets with the tagged sentiment, as well as my own manually created dataset, and finally, a dataset that was manually created via looking at following comments (for instance, "no need to be negative", is used to tag the prior comment.
The probable current mood is a prediction of your prior moods. I believe right now, it looks at the last time you discussed the given topic you searched (if you searched for a user, it is the most recent mood). If there is enough data, it'll also predict your mood, based on the time of day + topic being discussed.
Finally, I should add - there are definitely improvements to be made, and I'm currently working on mood. Primarily, the focus was on expertise, knowledge, and interests. That is what enables the "related content" section and is now very accurate (given a decent number of data samples). Mood, as stated, is ~80% accurate based on my tags. I think I can do better (probably >90%), especially with more labeled data for this particular dataset.
The goal of this is it can drop into any company and instantly can search for who and what is relevant to your company.
Thanks for sharing these details. I'd like to chat more about it if you'll bear with me:
Last I checked I was "tense". Is it theoretically possible to explain why your models came to that conclusion? Is there a theoretical data view that could say "here's the comments that contributed to you being considered "tense" (or even better: specific parts of the comments).
From my perspective, without a way to concretely back up any conclusion made, I simply cannot consider the complex-sounding systems to be any better than RNG.
I alluded to this in my first comment, but I think it's imperative that if you're going to explicitly call out what mood you think someone was/is/will be, you need to back it up with "why" so that, at the very least, someone can say, "that's clearly an unfair representation of my mood and here's where the algorithms messed up."
I'm going to go on a bit of a limb and assume that you're just having some fun and exploring an interesting feature. And I think that's really quite cool. But I think we're getting into a software engineering ethics realm that I'm ill-equipped to dig in to. You've got a site that applies "moods" to individuals in a way that doesn't appear to be verifiable. And even if they are, and are accurate, is that even the right thing to do?
I don't want to ruin your fun or twist your arm into changing this. I'm just responding to my red-alert alarm that lives in my stomach. It's set off by the discovery that one of my few online accounts where I don't hide my identity has been assigned a mood label.
First, this conversation is not going to be comforting and your red-alert alarm should be going off. I know this is building toward the dystopian future. I personally use Signal, put my cell phone in the microwave when I come home, and am writing this through a VPN.
That being said, I'm 100% confident that Google, Apple, Amazon, and more know WAY more about you than I do. This system only looks at public data and only links back to usernames. Strangely enough, it can identify users across sites (even without knowing your identity) - say Reddit and Hacker News. It was originally built for https://projectpiglet.com/ to find company insiders, and make trades (which works ridiculously well btw).
There should be no expectation of privacy.
> You've got a site that applies "moods" to individuals in a way that doesn't appear to be verifiable. And even if they are, and are accurate, is that even the right thing to do?
The short answer is, I don't know. I don't want this to exist, but it does. I also know that I have already built systems for companies doing this and more. Albeit the system here is much more of a "turn-key" solution and its patent-pending for what it's worth (I'm also against patents and donate to the EFF, but feel I can potentially segment off this market). Still, lots of people are looking to build systems such as this. Personally, I'd rather be the one guiding the ship, because I do ask myself the same questions you're asking.
I'll be honest... I've used this system to show my friends that even when they change usernames I can still identify them. That particular system (which looks at speech patterns) is turned off here. That's in part because of those questions. However, I'm 100% confident, others are doing that right now, with this conversation. We know that because Snowden shared it with us.
> Strangely enough, it can identify users across sites (even without knowing your identity) - say Reddit and Hacker News.
I think a larger demonstration could make for a very meaningful "Show HN" with hopefully a large impact on privacy consciousness.
I'm sure some of us expected this, but to have it demonstrated by a single person in a casual "Oh, it could do this as well, but I don't have it turned on for ethical reasons" manner is quite effective.
Maybe it will raise awareness that "big data" isn't just trying to correlate cat-pictures one posts on their communication medium to cat food advertising, but everything you write anywhere on the Internet, even on different media under different (or no) accounts, into permanent and guarded profiles with no recourse of opting-out of the machine, effectively rendering informal discourse over the Internet dead.
This is very odd. How did it happen that you're the ship-steerer for this highly cutting-edge tech? Where did the research come from? Usually, this kind of stuff doesn't just pop out of the ether, and I've never seen anything quite so promising in this domain before. (Maybe I just haven't been paying attention.)
> Personally, I'd rather be the one guiding the ship, because I do ask myself the same questions you're asking.
I'm not necessarily the one guiding the ship, but I'd like to be (by being the largest on the market), and I'm just working to launch this as a business (called Metacortex). Fact is, I've spent years building this out for my trading platform. Most of the research is my own, with some contributions from open source and academia as it relates to NLP. That's why I submitted a patent on it, I want to corner this particular (and seemingly highly effective) method.
That being said, I've worked on products with similar goals, and have done technical reviews of WAY creepier products. Luckily, those other products rarely work at all. IMO the technology isn't quite there yet (outside of some of what I demoed), however, it's uncomfortably close.
Snowden told us this was happening, years ago. I'm sure they've improved since.
> That's why I submitted a patent on it, I want to corner this particular (and seemingly highly effective) method.
Which (turns out one of my areas of interest is "which") really only gives you a monopoly on selling the method but not on internal usage by less scrupulous actors...at least that's my opinion, it would be near impossible to enforce a patent on non-public facing code used to track people in BigCorp.
I do like the idea though, it's not like anyone (reasonably) should believe they have any semblance of anonymity on the internets and this just goes to show how easy it is to follow you around on the webs. If an individual can pull this off then the sky's the limit for the TLAs with massive budgets and computing horsepower.
lol And now, I'll promptly shut down that system again...
As I said in the previous comment, that IMO is going too far. Even though I'm 100% sure others are using similar systems, I don't feel comfortable in that business -_-
Kinda highlights, although the hnprofile.com demo may have it's current faults - there's a lot you can do. The cool part, is the platform (called Metacortex) is easy to build apps on top of.
Fascinating! Have you considered using the tool to try identify who wrote the recent NY Times anonymous piece? (Assuming you can find enough source material from whitehouse staff)
Hey, I’m in Chi too! Right by North and Clybourn’s redline stop. If you ever want to reverse engineer this together or figure out how to train a drone to fly itself, drop an email / keybase.
...so...you feel tense? I feel tense reading this. But yes, I wouldn't like to read that either.
Ok, now I've waded in, I should do something to help. Reads the first few pages of your comments.. hmm Nah, you don't seem tense. All kinds of emotions. It's not you, it's HNProfile. Tries to think why it might say that.. There were a few with emotional words, frustration with situations..probably slightly more negative-emotion words than positive.. hmm I think maybe you just write better than most people, more vividly, mostly about serious topics. And are engaged with them. I don't know if it helps, but I say That's clearly an unfair representation of your mood. (I don't know you, wasn't paid by you)
I have an even better way to identify who is an expert on HN. Say "It's silly that X still isn't a thing." If X is a thing, you'll get ten people telling you why you're wrong within ten minutes. Example: https://news.ycombinator.com/item?id=17941772
You'll have to trade some karma for the answer though.
What else is karma for? :D
I have seen this technique used to great effect inside large companies too. People will spend a lot of effort telling you why you are dead wrong.
This is neat! But, it seems a bit disingenuous to say I'm looking at experts, as opposed to people who like to talk about something. My expertise is industrial automation and connectivity for example, but I come up when searching for vue, a javascript framework, just because it's a tool I like to promote since I think it could use some noise given react's popularity. Step 2 seems like having someone or something reading through comments and assigning expert status to profiles.
For reference, this started as a "different kind of search engine", utilizing conversation.
Mood wasn't exactly the focus, so it's robustness is limited - i.e. lots of room for improvement. The mood detection (based on my manual tagging) is roughly 80% accurate, which IMO was "good enough". It's more accurate as you get to the edges (very positive or very negative), the middle is kinda fuzzy, but that's kind of the nature of language.
Funny, I guess I am an expert in "heck". There is a certain rightness to that given the last couple of weeks. Is there a way to see the comments it deems 'happy'?
[edit] I do wonder if the analysis of topics is affected by both the frequency something shows up on HN[1] and not repeating the keyword in a reply post.
1) less examples and more likely to miss a topic a person is expert on
That's because it doesn't blacklist words which are commonly used in English without them being technical terms, nor doesn't it whitelist terms which are topics of interest. There's also no difference being made between referring to Wikipedia and being an expert about Wikipedia. If you regularly reference to Wikipedia (something I do) you then become an expert on the topic "wikipedia". As an example, check the expert topics from user tptacek [1] ("security, argument, fact, hn, thread, software, comment, story, google, sense").
The probable mood was rather... entertaining.
The notion this can be used to find sockpuppets or throwaways does not surprise me. Without sufficient effort you still use certain words, in certain order, consistently. Such can even be used to check out what your native language is. So if you are e.g. a blackhat who wrote ransomware it makes sense to have someone who never wrote on the internet to paraphrase or deal with the communication, or perhaps use tools such as translators.
Also, if you use delay (e.g. delay 10) it is going to appear on HN after you wrote it meaning it could appear in the next hour despite being written in the previous.
Despite the above, I did find this an interesting project; I just very much doubt its accuracy and usefulness.
The opt out currently, for privacy moving forward, is to cease contribution to the digital noosphere.
I suppose if you desired you could consolidate your doxx point to Google and ISP's by running all your comments through a couple rounds of Google translate.
There is no privacy possible in the global village. Expectation of such is technically naive.
I see there's no privacy policy, and no clear guidelines on opting out. Why? Privacy was the focal point of 2018, but you're somehow ignoring that altogether?
You can identify who is an expert related to each tool or any topic in general