-- B L A C K F R I D A Y -- by LETs #1 TOP PROVIDER RackNerd! - HUNDREDS OF GIVEAWAYS! COME GET SOME

Xrmaddness · November 2023

@DeadlyChemist said:

@Xrmaddness said:

@DeadlyChemist said:

@codelock said:
🥑🍞🫔🥯🥑

@codelock said:
💦🍆😁🥲😘

@codelock said:
⛵🚟🚡🇰🇿♾️

this is why we need a filter

Okay, now this I can certainly get behind. I have a hard time trying to justify how this follows the note from the participation post:

For posts to be eligible and counted -- they should be generally helpful and neighborly towards the community.

https://lowendtalk.com/discussion/comment/3753740/#Comment_3753740

3rd bf here
never has anyone been disqualified
either way it's a bad situtaion for all of us

Same here! Looks like we joined LET around the same time. I know the threads have had this same discussion pretty much every time, at some point, and, like you've mentioned, no one to my knowledge has ever been disqualified. I'm sure action would be taken if anything ever became too ridiculous.

Dustin does have a stipulation in the OP:

It’s our goal for this to be fun for everyone and it’s community-focused, meaning we’re all among friends - therefore we reserve the right to disregard posts that are naughty or just not nice!

So obviously it's up to his discretion.

codelock · November 2023

Towards page 200

Xrmaddness · November 2023

Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. Remember to try to keep it fun, that's what this is all about. Well, that and RackNerd services

Take care

MrEd · November 2023

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

No, the code is on my private SVN (I know, its not GIT, I am old enough to use old technologies )

But for this exact problem its easy:

private LevenshteinDistance distance = new LevenshteinDistance();

...

int dist = userComments.stream().map(uc -> distance.apply(uc.getBody(), commentBody)).min(Integer::compareTo).orElse(Integer.MAX_VALUE);
double similarity = (commentBody.length() - dist) / (double) commentBody.length();
if (similarity > 0.9d) {
  counter.setSimilar(counter.getSimilar() + 1);
}

I already mentioned, that I am using LevenshteinDistance from apache common-text library.

DeadlyChemist · November 2023

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

i'd include ai to verify the messages....

DeadlyChemist · November 2023

@codelock said:
😵😣😰😵🤡🥸

@codelock said:
Page 197 welcome

i feel like he's a bot

codelock · November 2023

@DeadlyChemist said:

@codelock said:
😵😣😰😵🤡🥸

@codelock said:
Page 197 welcome

i feel like he's a bot

Nope I am not

codelock · November 2023

All emojis have some hidden meaning but none of you @FrankZ are trying to find it

DeadlyChemist · November 2023

@MrEd said:

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

No, the code is on my private SVN (I know, its not GIT, I am old enough to use old technologies )

But for this exact problem its easy:
private LevenshteinDistance distance = new LevenshteinDistance();

...

int dist = userComments.stream().map(uc -> distance.apply(uc.getBody(), commentBody)).min(Integer::compareTo).orElse(Integer.MAX_VALUE);
double similarity = (commentBody.length() - dist) / (double) commentBody.length();
if (similarity > 0.9d) {
  counter.setSimilar(counter.getSimilar() + 1);
}
I already mentioned, that I am using LevenshteinDistance from apache common-text library.

im tempted to add AI to it, i have access to deepai their chatgpt model... need to check MY pricing but i think it''s affordable enough

markz · November 2023

I use a ====VPS RACKNERD==== gift party

codelock · November 2023

@MrEd said:

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

No, the code is on my private SVN (I know, its not GIT, I am old enough to use old technologies )

But for this exact problem its easy:
private LevenshteinDistance distance = new LevenshteinDistance();

...

int dist = userComments.stream().map(uc -> distance.apply(uc.getBody(), commentBody)).min(Integer::compareTo).orElse(Integer.MAX_VALUE);
double similarity = (commentBody.length() - dist) / (double) commentBody.length();
if (similarity > 0.9d) {
  counter.setSimilar(counter.getSimilar() + 1);
}
I already mentioned, that I am using LevenshteinDistance from apache common-text library.

How do you differentiate between images? A lot of images and gifs posted that might not always have same link

DeadlyChemist · November 2023

@codelock said:

@DeadlyChemist said:

@codelock said:
😵😣😰😵🤡🥸

@codelock said:
Page 197 welcome

i feel like he's a bot

Nope I am not

kay (not)

codelock · November 2023

@markz said:
I use a ====VPS RACKNERD==== gift party

Welcome to the party

codelock · November 2023

Discuss everything you want but even most useless comment pushes this thread to top so is helping them

DeadlyChemist · November 2023

@DeadlyChemist said:
@MrEd said:

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

No, the code is on my private SVN (I know, its not GIT, I am old enough to use old technologies )

But for this exact problem its easy:
private LevenshteinDistance distance = new LevenshteinDistance();

...

int dist = userComments.stream().map(uc -> distance.apply(uc.getBody(), commentBody)).min(Integer::compareTo).orElse(Integer.MAX_VALUE);
double similarity = (commentBody.length() - dist) / (double) commentBody.length();
if (similarity > 0.9d) {
  counter.setSimilar(counter.getSimilar() + 1);
}
I already mentioned, that I am using LevenshteinDistance from apache common-text library.
im tempted to add AI to it, i have access to deepai their chatgpt model... need to check MY pricing but i think it''s affordable enough

nvm i can spam ai picutres but the chat is expensive for my account

codelock · November 2023

😘🥸🤏

markz · November 2023

@codelock said:

@markz said:
I use a ====VPS RACKNERD==== gift party

Welcome to the party

Tynk, waiting hour by hour, but haven't been lucky yet, hahaha

MrEd · November 2023

@codelock said:
@MrEd said:

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

No, the code is on my private SVN (I know, its not GIT, I am old enough to use old technologies )

But for this exact problem its easy:
private LevenshteinDistance distance = new LevenshteinDistance();

...

int dist = userComments.stream().map(uc -> distance.apply(uc.getBody(), commentBody)).min(Integer::compareTo).orElse(Integer.MAX_VALUE);
double similarity = (commentBody.length() - dist) / (double) commentBody.length();
if (similarity > 0.9d) {
  counter.setSimilar(counter.getSimilar() + 1);
}
I already mentioned, that I am using LevenshteinDistance from apache common-text library.
How do you differentiate between images? A lot of images and gifs posted that might not always have same link

If the links is different, it is considered as different content, I don't parse the images themselves

codelock · November 2023

@DeadlyChemist said:
@DeadlyChemist said:
@MrEd said:

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

No, the code is on my private SVN (I know, its not GIT, I am old enough to use old technologies )

But for this exact problem its easy:
private LevenshteinDistance distance = new LevenshteinDistance();

...

int dist = userComments.stream().map(uc -> distance.apply(uc.getBody(), commentBody)).min(Integer::compareTo).orElse(Integer.MAX_VALUE);
double similarity = (commentBody.length() - dist) / (double) commentBody.length();
if (similarity > 0.9d) {
  counter.setSimilar(counter.getSimilar() + 1);
}
I already mentioned, that I am using LevenshteinDistance from apache common-text library.
im tempted to add AI to it, i have access to deepai their chatgpt model... need to check MY pricing but i think it''s affordable enough
nvm i can spam ai picutres but the chat is expensive for my account

I invite you to post them 24*7 without boting your account, for entire month hopefully you don't have other things in your life

MrEd · November 2023

@DeadlyChemist said:
@MrEd said:

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

No, the code is on my private SVN (I know, its not GIT, I am old enough to use old technologies )

But for this exact problem its easy:
private LevenshteinDistance distance = new LevenshteinDistance();

...

int dist = userComments.stream().map(uc -> distance.apply(uc.getBody(), commentBody)).min(Integer::compareTo).orElse(Integer.MAX_VALUE);
double similarity = (commentBody.length() - dist) / (double) commentBody.length();
if (similarity > 0.9d) {
  counter.setSimilar(counter.getSimilar() + 1);
}
I already mentioned, that I am using LevenshteinDistance from apache common-text library.
im tempted to add AI to it, i have access to deepai their chatgpt model... need to check MY pricing but i think it''s affordable enough

It would be interesting to try to do something with AI, but that would be out of scope of "quick and dirty" changes Already half of my work day has passed and I didn't find a timeframe to make small changes I already intended

codelock · November 2023

@markz said:

@codelock said:

@markz said:
I use a ====VPS RACKNERD==== gift party

Welcome to the party

Tynk, waiting hour by hour, but haven't been lucky yet, hahaha

@dustinc becomes active around 8 pm ist , there are also giveaways during 7 am ist but right now he is probably sleeping

codelock · November 2023

Time mentioned is in ist aka Indian standard time don't know what time that is at other places

DeadlyChemist · November 2023

@codelock said:
@DeadlyChemist said:
@DeadlyChemist said:
@MrEd said:

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

No, the code is on my private SVN (I know, its not GIT, I am old enough to use old technologies )

But for this exact problem its easy:
private LevenshteinDistance distance = new LevenshteinDistance();

...

int dist = userComments.stream().map(uc -> distance.apply(uc.getBody(), commentBody)).min(Integer::compareTo).orElse(Integer.MAX_VALUE);
double similarity = (commentBody.length() - dist) / (double) commentBody.length();
if (similarity > 0.9d) {
  counter.setSimilar(counter.getSimilar() + 1);
}
I already mentioned, that I am using LevenshteinDistance from apache common-text library.
im tempted to add AI to it, i have access to deepai their chatgpt model... need to check MY pricing but i think it''s affordable enough
nvm i can spam ai picutres but the chat is expensive for my account
I invite you to post them 24*7 without boting your account, for entire month hopefully you don't have other things in your life

check previous posts to understand it
also, says a guy spamming emojis lol

codelock · November 2023

😹😹😹😹

DeadlyChemist · November 2023

@MrEd said:
@DeadlyChemist said:
@MrEd said:

@Xrmaddness said:
Anyway, I'm off to bed. I hope you all have a wonderful day or night, whichever it is currently wherever you are. > @MrEd said:

@Xrmaddness said: At any rate, did I miss a post where you explained or even shared the exact code that you're using to determine the difference between a unique and non-unique post? I'd say that's quite important for anyone to determine for themselves if your tracking is even valid.

>

@MrEd said:

@Arjun42 said:
Thanks! Hopefully it's easier to deal with than Levenshtein Distance.

I will though implement some other "similarity" measure, because currently it is 90% according to the length of the message... This doesn't fire on messages "Page 123" and "Page 124", because difference is 1 for these messages, they are clearly similar, but because message itself is short (8 characters), similarity is only 87.5%, and it doesn't fall into similar messages category. I guess I will have to add a check "if difference is less than 3" for example, then ignore the percentage This will definitely increase the number of similar comments, maybe we will have less 0% members

I did explain it here. I hope it is clear enough how it works, but as mentioned in the comment, it has its flaws which I will try to get around (maybe with distance < 3 == similar)

Thanks for drudging up your previous post, I had figured you already explained it, but there are far too many pages. I appreciate it, thank you.

I'm not sure if you share your code publicly anywhere, but I'd love to have a look, at least for this part of it. I'm just genuinely curious, it's an interesting problem to solve. It sounds like you're still making adjustments to it?

No, the code is on my private SVN (I know, its not GIT, I am old enough to use old technologies )

But for this exact problem its easy:
private LevenshteinDistance distance = new LevenshteinDistance();

...

int dist = userComments.stream().map(uc -> distance.apply(uc.getBody(), commentBody)).min(Integer::compareTo).orElse(Integer.MAX_VALUE);
double similarity = (commentBody.length() - dist) / (double) commentBody.length();
if (similarity > 0.9d) {
  counter.setSimilar(counter.getSimilar() + 1);
}
I already mentioned, that I am using LevenshteinDistance from apache common-text library.
im tempted to add AI to it, i have access to deepai their chatgpt model... need to check MY pricing but i think it''s affordable enough
It would be interesting to try to do something with AI, but that would be out of scope of "quick and dirty" changes Already half of my work day has passed and I didn't find a timeframe to make small changes I already intended

true true

codelock · November 2023

https://media.tenor.com/tXmPJApPL5UAAAAM/taylor-swift-tay.gif

Tay tay is back

codelock · November 2023

codelock · November 2023

Nobodywelcomed new page 198

codelock · November 2023

😰😰😰😣🥲

codelock · November 2023

♾️♾️♾️🇰🇿

Howdy, Stranger!

Categories

In this Discussion

-- B L A C K F R I D A Y -- by LETs #1 TOP PROVIDER RackNerd! - HUNDREDS OF GIVEAWAYS! COME GET SOME

Comments

Howdy, Stranger!

Quick Links

Categories

In this Discussion

-- B L A C K F R I D A Y -- by LETs #1 TOP PROVIDER RackNerd! - HUNDREDS OF GIVEAWAYS! COME GET SOME

Comments