Title: Trust List Analysis - Alt Account Identification Post by: dogie on October 03, 2015, 05:30:24 PM Following on from the Trust list visualisations thread (https://bitcointalk.org/index.php?topic=1062052.0) and the ethics consultation thread (https://bitcointalk.org/index.php?topic=1199149), I've decided to release the analysis of a tool which may be able to identify multiple accounts which belong to the same owner. While some of the results are statistically significant, this should still only constitute one piece of evidence of a link between accounts.
The methodology is less important than the output, which can be confirmed and visualised manually.
Along with potentially linked accounts, the list below has 2 numerical metrics attached:
Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 03, 2015, 05:30:28 PM Below is a highly filtered list of the first batch of information. Notes:
Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 03, 2015, 05:30:34 PM This list is considered low probability. It really doesn't mean much without a lot of other information in order to confirm a link.
Visualisation of the 1st and 2nd batches of names:
https://i.imgur.com/tphYSQT.png (https://i.imgur.com/hUUZpu4.png) Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 03, 2015, 05:30:48 PM Reserved
Title: Re: Trust List Analysis - Alt Account Identification Post by: Lauda on October 03, 2015, 05:42:35 PM I won't post in the other thread to say that I support this (since it would be a useless post) and you've already posted the tool. The analysis and system used is actually very interesting to say at least. This might also be the tipping stone in some cases where it was already 'probable' that X was alt of Y.
Quote Trust lists with only "DefaultTrust" (unedited) are not included Let me get this straight, so users without edited lists are completely excluded? I used to have an alt (I doubt that anyone besides an admin even knew about it) with a unedited list. This means that you could to identify that account as my probable alt, correct?Title: Re: Trust List Analysis - Alt Account Identification Post by: Shield on October 03, 2015, 05:51:24 PM The question is how accurate this analysis is?
it may get some legit users in problem also, how it was done and what is its accuracy Title: Re: Trust List Analysis - Alt Account Identification Post by: Xian01 on October 03, 2015, 05:56:40 PM
I miss those Butterfly Labs folks... I wonder if any of them has been hit by a bus / random falling space debris, or have come down with some painful incurable affliction since they left the forums. http://ia802308.us.archive.org/32/items/gov.uscourts.mowd.117531/gov.uscourts.mowd.117531.docket.html bcp19, Inaba, and SLok, y u no post any more ? :( Title: Re: Trust List Analysis - Alt Account Identification Post by: bitcoin revo on October 03, 2015, 05:57:01 PM The question is how accurate this analysis is? it may get some legit users in problem also, how it was done and what is its accuracy It's simply a tool. I doubt that any real alt-finding will be on the sole basis of this proof; it'll be more like another small evidence bit that accompanies more substantial proof. I won't post in the other thread to say that I support this (since it would be a useless post) and you've already posted the tool. The analysis and system used is actually very interesting to say at least. This might also be the tipping stone in some cases where it was already 'probable' that X was alt of Y. -snip- Let me get this straight, so users without edited lists are completely excluded? I used to have an alt (I doubt that anyone besides an admin even knew about it) with a unedited list. This means that you could to identify that account as my probable alt, correct? From what I understand, if your alt has an unedited list (just DefaultTrust in it), then the tool will exclude your alt. Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 03, 2015, 06:07:50 PM I won't post in the other thread to say that I support this (since it would be a useless post) and you've already posted the tool. The analysis and system used is actually very interesting to say at least. This might also be the tipping stone in some cases where it was already 'probable' that X was alt of Y. Yeah I think that's the main use of this. "Hmm just scambusted that guy, lets see if he could still be floating about".Let me get this straight, so users without edited lists are completely excluded? I used to have an alt (I doubt that anyone besides an admin even knew about it) with a unedited list. This means that you could to identify that account as my probable alt, correct? I could NOT identify your alt if it had just "DefaultTrust" in its trust list. Accounts with an unedited trust list aren't included in the trustlist dump Theymos provides.The question is how accurate this analysis is? As I said in the other thread, the methodology isn't important on this one as you can do a human check across the two trust lists (I've added a shortened version alongside). Its as accurate as you want it to be and could be entirely useless in some cases but pretty strong in others.it may get some legit users in problem also, how it was done and what is its accuracy Title: Re: Trust List Analysis - Alt Account Identification Post by: onemorexmr on October 03, 2015, 06:09:55 PM got me ;)
onemorebtc is an alt of me. but i dont use it for posting anymore Title: Re: Trust List Analysis - Alt Account Identification Post by: smoothie on October 03, 2015, 06:16:51 PM Interesting and watching
Title: Re: Trust List Analysis - Alt Account Identification Post by: dooglus on October 03, 2015, 06:44:44 PM Along with potentially linked accounts, the list below has 2 numerical metrics attached:
Do you consider incoming trust as well as outgoing? If someone is 'farming' trust, they will probably be leaving positive trust on multiple alts at a time. Recognising such patterns could maybe make your tool stronger. Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 03, 2015, 07:00:14 PM Do you consider incoming trust as well as outgoing? If someone is 'farming' trust, they will probably be leaving positive trust on multiple alts at a time. Recognising such patterns could maybe make your tool stronger. No it doesn't although its worth remembering this is trust lists, not trust ratings. Being trusted by multiple people is hard to factor in although it could be used as an end stage check to see if accounts are cross trusting each other. For the time being I'll be working on visualisation and automation of the interpretation so I don't have to manually review 1,000s of them. Title: Re: Trust List Analysis - Alt Account Identification Post by: !! pop on October 03, 2015, 07:03:48 PM re. "Data isn't being spat out in a nice form so formatting will look meh for now. I'll add in some of my other table codes in time."
You mentioned using Excel to generate this? If so, could you dump/export this to Google Docs, formulas & all? Title: Re: Trust List Analysis - Alt Account Identification Post by: shorena on October 03, 2015, 07:44:35 PM Just to make sure we are on the same page. This is a measure how similar peoples trust lists are as well as a measure how large (as in complex) the average trust list between the two is. complexity(A,B) := [Count(A)*.3 + Count(B)*.3] / 2
It does however not - assuming here, please correct me if Im wrong - take into account how likely it is for someone to end up on a given trust list in the first place. From the trust.txt I downloaded just a few minutes ago I see that - as expected - certain users are more often on someones list than others. You have a top list for this as well IIRC in your other thread. E.g. theymos is on 170 different trust lists, dogie on 34. I also cant follow the premise that a person would use an identical (or very similar) trust list over multiple accounts. To sum it up, what makes you draw the conclusion that similarity of trust lists has anything to do with alt accounts? Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 03, 2015, 08:20:15 PM re. "Data isn't being spat out in a nice form so formatting will look meh for now. I'll add in some of my other table codes in time." You mentioned using Excel to generate this? If so, could you dump/export this to Google Docs, formulas & all? I meant that the final data the analysis spits out isn't neccesarily interpretable or useful for humans straight away and requires a significant amount of post processing. Turning it from Excel into forum code isn't a problem, I have plenty of translators from my PSU / company table based guides I can reuse. There's quite a lot of manual work and some Excel only add-ons which would prevent Google Doc replication, but results could be published there. I'm still reluctant to dump a raw list without forcing a 'reasonable' interpretation on the information, I don't want this list to be weaponised. Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 03, 2015, 08:20:25 PM Just to make sure we are on the same page. This is a measure how similar peoples trust lists are as well as a measure how large (as in complex) the average trust list between the two is. complexity(A,B) := [Count(A)*.3 + Count(B)*.3] / 2 The closest thing you'll find to this is the Jaccard index.Quote You take two sample sets. You count the attributes shared by both samples (call this Z). You count the attributes observed in only one sample (call this X). You then count the attributes observed in only the other sample (call this Y). Finally, you can calculate the similarity index by using the three counts in this operation: Z / (Z+X+Y). It does however not - assuming here, please correct me if Im wrong - take into account how likely it is for someone to end up on a given trust list in the first place. From the trust.txt I downloaded just a few minutes ago I see that - as expected - certain users are more often on someones list than others. You have a top list for this as well IIRC in your other thread. E.g. theymos is on 170 different trust lists, dogie on 34. It does not, that's why I'm manually screening anything I release at the moment, there are plenty of "strong correlation" results I filtered out because they're common results. Depending on how much time I invest in this, I may filter out some of the top people altogether.I also cant follow the premise that a person would use an identical (or very similar) trust list over multiple accounts. They do, because for anyone not in DefaultTrust (99% of people), the only utilisation for their trust list is to show who's ratings they want to see. And regardless of which account they're using those preferences generally remain constant.To sum it up, what makes you draw the conclusion that similarity of trust lists has anything to do with alt accounts? Past experience.Title: Re: Trust List Analysis - Alt Account Identification Post by: GermanGiant on October 03, 2015, 11:19:21 PM
I also cant follow the premise that a person would use an identical (or very similar) trust list over multiple accounts. They do, because for anyone not in DefaultTrust (99% of people), the only utilisation for their trust list is to show who's ratings they want to see. And regardless of which account they're using those preferences generally remain constant.To sum it up, what makes you draw the conclusion that similarity of trust lists has anything to do with alt accounts? Past experience.Title: Re: Trust List Analysis - Alt Account Identification Post by: BitcoinEXpress on October 03, 2015, 11:39:30 PM Channeling my inner Techshare
Okay so now you've basically proven there are people abusing the DT, what's next? Are you going to next tell us that people are wrongfully removed from DT because they speak out against injustice? closing Techshare channel ~BCX~ P.S. Excellent work Dogie. Title: Re: Trust List Analysis - Alt Account Identification Post by: shorena on October 04, 2015, 04:08:09 PM Just to make sure we are on the same page. This is a measure how similar peoples trust lists are as well as a measure how large (as in complex) the average trust list between the two is. complexity(A,B) := [Count(A)*.3 + Count(B)*.3] / 2 The closest thing you'll find to this is the Jaccard index.Quote You take two sample sets. You count the attributes shared by both samples (call this Z). You count the attributes observed in only one sample (call this X). You then count the attributes observed in only the other sample (call this Y). Finally, you can calculate the similarity index by using the three counts in this operation: Z / (Z+X+Y). Thanks for that. It does however not - assuming here, please correct me if Im wrong - take into account how likely it is for someone to end up on a given trust list in the first place. From the trust.txt I downloaded just a few minutes ago I see that - as expected - certain users are more often on someones list than others. You have a top list for this as well IIRC in your other thread. E.g. theymos is on 170 different trust lists, dogie on 34. It does not, that's why I'm manually screening anything I release at the moment, there are plenty of "strong correlation" results I filtered out because they're common results. Depending on how much time I invest in this, I may filter out some of the top people altogether.Which could be argued to bring in bias since it relies on your perspective and past experiences. I also cant follow the premise that a person would use an identical (or very similar) trust list over multiple accounts. They do, because for anyone not in DefaultTrust (99% of people), the only utilisation for their trust list is to show who's ratings they want to see. And regardless of which account they're using those preferences generally remain constant.To sum it up, what makes you draw the conclusion that similarity of trust lists has anything to do with alt accounts? Past experience.It did not for Quickseller and the (now known) alt, just to name one example of someone with a hidden alt and a highly modified trust list. There are different ways to get the same result and only the direct copy would be noticable by your reasearch. Additionally for those that actively try to hide their alts its very easy to manipulate their list in such a way that they no longer end up over 0.8 or any other limit for that matter. Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 04, 2015, 04:57:08 PM It does however not - assuming here, please correct me if Im wrong - take into account how likely it is for someone to end up on a given trust list in the first place. From the trust.txt I downloaded just a few minutes ago I see that - as expected - certain users are more often on someones list than others. You have a top list for this as well IIRC in your other thread. E.g. theymos is on 170 different trust lists, dogie on 34. It does not, that's why I'm manually screening anything I release at the moment, there are plenty of "strong correlation" results I filtered out because they're common results. Depending on how much time I invest in this, I may filter out some of the top people altogether.There are different ways to get the same result and only the direct copy would be noticable by your reasearch. Additionally for those that actively try to hide their alts its very easy to manipulate their list in such a way that they no longer end up over 0.8 or any other limit for that matter. The more you ask about the algorithm the easier it is to game, but as you'll see in later releases its no where near as easy as you're making it out to be to mask your list if you want it to be functional. The last batch I processed last night used ratings as low as 0.5 and there are plenty of strong correlations down there, so a future run will take even more. I also set maximum matches to 5 thinking most of these guys wouldn't have more than 5 but there are plenty already hitting that boundary.Title: Re: Trust List Analysis - Alt Account Identification Post by: Quickseller on October 04, 2015, 05:33:02 PM It did not for Quickseller and the (now known) alt, just to name one example of someone with a hidden alt and a highly modified trust list. I don't see any reason to use more then one trust list across multiple accounts, at least no legit reason. (one possible reason would be to make someone look more trustworthy or less trustworthy then they really are). If I want to see someone's trust score using QS's trust list then I will look at their profile from QS, and if I want to see their trust score with DefaultTrust then I will look at their profile from an alt. There are different ways to get the same result and only the direct copy would be noticable by your reasearch. Additionally for those that actively try to hide their alts its very easy to manipulate their list in such a way that they no longer end up over 0.8 or any other limit for that matter. The only time that I would modify the trust list of an alt account would be for testing purposes to look at someone's trust list/trust network. This is now somewhat of a moot reason because the entire trust network is published every Friday night (EST, Saturday morning GMT). This kind of research could be used to monitor for scammers who are attempting to farm trust, as many trust farmers tend to ask for both trust feedback and to be added to other's trust lists. Title: Re: Trust List Analysis - Alt Account Identification Post by: shorena on October 04, 2015, 05:39:29 PM -snip- If 2 people both have Theymos, John K and BadBear in their lists, the correlation would be 100% but it doesn't tell us anything about the relationship between the two accounts because its so common. The current version doesn't weight based on frequency yet so it needs manual filtering.Which could be argued to bring in bias since it relies on your perspective and past experiences. I agree that some additions to a trust list might be common while others are not, my point was that it brings bias if you alone decide it what is common and what not. There are different ways to get the same result and only the direct copy would be noticable by your reasearch. Additionally for those that actively try to hide their alts its very easy to manipulate their list in such a way that they no longer end up over 0.8 or any other limit for that matter. The more you ask about the algorithm the easier it is to game, but as you'll see in later releases its no where near as easy as you're making it out to be to mask your list if you want it to be functional. The last batch I processed last night used ratings as low as 0.5 and there are plenty of strong correlations down there, so a future run will take even more. I also set maximum matches to 5 thinking most of these guys wouldn't have more than 5 but there are plenty already hitting that boundary.What if I told you that I did not post this yesterday because I was pondering whether or not I should post how to easily game your analysis, but still see identical trust ratings? I have not gambled for a while, would you be willing to make it a small bet? Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 04, 2015, 06:02:04 PM -snip- If 2 people both have Theymos, John K and BadBear in their lists, the correlation would be 100% but it doesn't tell us anything about the relationship between the two accounts because its so common. The current version doesn't weight based on frequency yet so it needs manual filtering.Which could be argued to bring in bias since it relies on your perspective and past experiences. There are different ways to get the same result and only the direct copy would be noticable by your reasearch. Additionally for those that actively try to hide their alts its very easy to manipulate their list in such a way that they no longer end up over 0.8 or any other limit for that matter. The more you ask about the algorithm the easier it is to game, but as you'll see in later releases its no where near as easy as you're making it out to be to mask your list if you want it to be functional. The last batch I processed last night used ratings as low as 0.5 and there are plenty of strong correlations down there, so a future run will take even more. I also set maximum matches to 5 thinking most of these guys wouldn't have more than 5 but there are plenty already hitting that boundary.What if I told you that I did not post this yesterday because I was pondering whether or not I should post how to easily game your analysis, but still see identical trust ratings? I have not gambled for a while, would you be willing to make it a small bet? Title: Re: Trust List Analysis - Alt Account Identification Post by: shorena on October 04, 2015, 07:10:34 PM -snip- -snip- I don't really understand your point. This analysis isn't meant to be the "no one can have an invisible ult account ever again" system, its simply used to unearth several 100 of them already in existence. Most of which are used for evil purposes.What if I told you that I did not post this yesterday because I was pondering whether or not I should post how to easily game your analysis, but still see identical trust ratings? I have not gambled for a while, would you be willing to make it a small bet? Maybe I just expected more than similarities between trust lists. Its certainly interesting work, I just dont think it can do what you suggest. On the other hand I might not have the background to judge the significance. Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 04, 2015, 07:36:57 PM -snip- -snip- I don't really understand your point. This analysis isn't meant to be the "no one can have an invisible ult account ever again" system, its simply used to unearth several 100 of them already in existence. Most of which are used for evil purposes.What if I told you that I did not post this yesterday because I was pondering whether or not I should post how to easily game your analysis, but still see identical trust ratings? I have not gambled for a while, would you be willing to make it a small bet? Maybe I just expected more than similarities between trust lists. Its certainly interesting work, I just dont think it can do what you suggest. On the other hand I might not have the background to judge the significance. Its an analysis of trust lists, its not magical :P That being said if I had access to other forum data I'm sure it could get exponentially more accurate. Have a look at some of the batch 2's I've released, lots of alts. Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 04, 2015, 08:31:01 PM Added batch 2, including some lower tier results. These should generally be ignored in isolation although can often identify some outliers to an existing alt network. Also added a visualisation of the first 2 batches of results.
Title: Re: Trust List Analysis - Alt Account Identification Post by: Blazr on October 05, 2015, 12:50:02 PM
Hmm I dont know about that... I don't think bbit was bought. I was the one who outed hashie originally and most evidence I've seen points to hashie and bbit being two separate individuals. Glancing through this list I see a few other accounts that I know for sure aren't alts. Nice idea but it isn't very accurate at all. Maybe you aren't thinking about this the right way, take a look at this one:
I would think that ahmed_bodi and mrbodi would more likely be alts of ahmedbodi rather than VERUMinNUMERIS. Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 05, 2015, 01:41:11 PM Hmm I dont know about that... I don't think bbit was bought. I was the one who outed hashie originally and most evidence I've seen points to hashie and bbit being two separate individuals. If they are 2 separate individuals then they are associated, which is what this analysis highlights.Glancing through this list I see a few other accounts that I know for sure aren't alts. Nice idea but it isn't very accurate at all. Glancing through the list will of course highlight names you say aren't alts, that doesn't mean the other 97% aren't associated in some way.Maybe you aren't thinking about this the right way, take a look at this one: You're looking in the really low association table which doesn't mean much. The higher association table already highlighted that ahmedbody = ahmed_bodi, and mrbodi isn't in the data because it has an unedited trust list. VERUMinNUMERIS is already strongly associated with 3 other accounts so its more than possible its the same person controlling them all.I would think that ahmed_bodi and mrbodi would more likely be alts of ahmedbodi rather than VERUMinNUMERIS. So in conclusion the analysis identified: 2 alt accounts with likely the same owner, 4 alt accounts with likely the same owner, and potentially a link between the two sets. I'd call that a success (see the diagram). Title: Re: Trust List Analysis - Alt Account Identification Post by: monkeynuts on October 05, 2015, 02:10:01 PM There is a danger here, that this analysis is identifying 'communities' within bitcoin rather than alts
for example - TookDK / Wheresmycoin Both I know well. Both I have bought and sold physical crypto coins from and to. Live on opposite sides of the world. Not alts. But are both keen physical crypto currency buyers / sellers / collectors. One example of a very niche community who buy / sell and trade with fellow members of the same niche community, that will by inference result in very common trust profiles. Reckon this is stating that the physical crypto bods need their own section. I guess there will be other similar examples in the lists. Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 05, 2015, 02:38:10 PM There is a danger here, that this analysis is identifying 'communities' within bitcoin rather than alts It may pick up trading communities, yes.for example - TookDK / Wheresmycoin That's just an error, I flagged them wrong at some point when comparing 1000s of lists. They probably wouldn't even qualify for the lower tier of association.Title: Re: Trust List Analysis - Alt Account Identification Post by: Lesbian Cow on October 05, 2015, 04:48:17 PM There is a danger here, that this analysis is identifying 'communities' within bitcoin rather than alts for example - TookDK / Wheresmycoin Both I know well. Both I have bought and sold physical crypto coins from and to. Live on opposite sides of the world. Not alts. But are both keen physical crypto currency buyers / sellers / collectors. One example of a very niche community who buy / sell and trade with fellow members of the same niche community, that will by inference result in very common trust profiles. Reckon this is stating that the physical crypto bods need their own section. I guess there will be other similar examples in the lists. CJBianco Lesbian Cow 0.61 6.02 I do not have any alts. Both of us do trade physical bitcoins - it is a fairly small community so it is not surprising we have similar trust lists. Title: Re: Trust List Analysis - Alt Account Identification Post by: TookDk on October 05, 2015, 05:16:01 PM There is a danger here, that this analysis is identifying 'communities' within bitcoin rather than alts It may pick up trading communities, yes.for example - TookDK / Wheresmycoin That's just an error, I flagged them wrong at some point when comparing 1000s of lists. They probably wouldn't even qualify for the lower tier of association.Yes, your algorithm got a false-positive on this one. Wheresmycoin is located 10000 Km from me, literately on the other side of the globe. Title: Re: Trust List Analysis - Alt Account Identification Post by: Quickseller on October 05, 2015, 06:05:15 PM There is a danger here, that this analysis is identifying 'communities' within bitcoin rather than alts for example - TookDK / Wheresmycoin Both I know well. Both I have bought and sold physical crypto coins from and to. Live on opposite sides of the world. Not alts. But are both keen physical crypto currency buyers / sellers / collectors. One example of a very niche community who buy / sell and trade with fellow members of the same niche community, that will by inference result in very common trust profiles. Reckon this is stating that the physical crypto bods need their own section. I guess there will be other similar examples in the lists. CJBianco Lesbian Cow 0.61 6.02 I do not have any alts. Both of us do trade physical bitcoins - it is a fairly small community so it is not surprising we have similar trust lists. I would vouch that this is a true statement, at least the part about CJBianco and Lesbian Cow being different people. I would not be surprised if many people in trading communities end up being shown as potentially the same person. Title: Re: Trust List Analysis - Alt Account Identification Post by: notlist3d on October 05, 2015, 09:39:21 PM Very nice work I really like the idea. And the method you used seems pretty sound, as some said a few false positives but I suspect you found a good amount of alt trust with this.
What tool did you use to do this? Did you make it yourself? Can you tell us some about it. I enjoy learning how things work. Title: Re: Trust List Analysis - Alt Account Identification Post by: bitcoin revo on October 05, 2015, 10:02:03 PM I would not be surprised if many people in trading communities end up being shown as potentially the same person. Well, the tool shouldn't really be used as anything more than an extra bit of evidence, IMHO. There are lots of possibilities that the "alts" were separate people who just had similar trusting personalities. Title: Re: Trust List Analysis - Alt Account Identification Post by: Mitchell on October 05, 2015, 10:06:40 PM I would be very surprised if CJBianco was an actual alt of Lesbian Cow (or vice versa).
Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 06, 2015, 09:30:35 AM Very nice work I really like the idea. And the method you used seems pretty sound, as some said a few false positives but I suspect you found a good amount of alt trust with this. What tool did you use to do this? Did you make it yourself? Can you tell us some about it. I enjoy learning how things work. Most things are in excel, yes I made it, no can't say too much about it but its based off the trust list data set Theymos published. Title: Re: Trust List Analysis - Alt Account Identification Post by: monkeynuts on October 06, 2015, 10:51:36 AM Very nice work I really like the idea. And the method you used seems pretty sound, as some said a few false positives but I suspect you found a good amount of alt trust with this. What tool did you use to do this? Did you make it yourself? Can you tell us some about it. I enjoy learning how things work. Most things are in excel, yes I made it, no can't say too much about it but its based off the trust list data set Theymos published. Can the outputs of your work be a feed into the work tspacepilot did ? ("Statistical Methods for Natural Language Processing"). Trust list identifies possible alt candidates, language processing gives probability ? Title: Re: Trust List Analysis - Alt Account Identification Post by: dogie on October 06, 2015, 11:18:08 AM Very nice work I really like the idea. And the method you used seems pretty sound, as some said a few false positives but I suspect you found a good amount of alt trust with this. What tool did you use to do this? Did you make it yourself? Can you tell us some about it. I enjoy learning how things work. Most things are in excel, yes I made it, no can't say too much about it but its based off the trust list data set Theymos published. Can the outputs of your work be a feed into the work tspacepilot did ? ("Statistical Methods for Natural Language Processing"). Trust list identifies possible alt candidates, language processing gives probability ? Yes. This really needs more data sources but they'll most likely have to come from Theymos. If you think this is hit and miss, language processing is 10x worse and would probably match everyone to everyone else. There is also no way to override or manually check it by human eyes as all it gives you is a name and a number. |