Advice for the website looksiq

[wais5psl]

I created the website LOOKSIQ because all existing systems are inherently flawed and viciously circular. Let’s take harmony as an example. Even in studies where ratios are combined with machine learning algorithms they explain only about 60% of the total variance and that’s using a cleaner and more scientific approach than what we typically see here where the scoring system is largely made up (for example 70% harmony= chadlite with no validation).

The modus operandi should not be simply attributing certain ratios to percentiles. Instead one can use ML algorithms like XGBoost to find the relationship between ratios and attractiveness examining both nonlinear and linear relationships. This is done by minimizing a loss function mostly multiclass classification loss by adjusting the weights of each feature in the model. The conclusion is that even near perfectly weighting and combining ratios does not produce an adequate measure of attractiveness.

The alternative I propose is Item Response Theory (IRT) which maps the relationship between the latent variable attractiveness and the likelihood that a person is perceived as more attractive than another person. If enough people vote individual biases cancel out under certain conditions according to the central limit theorem leaving only the true variance being the person’s “true” attractiveness.

Currently I am planning to add features so that users can give advice to others. This advice would then be standardized so that only advice aligned with the consensus of the general population remains ( meaning that most people would recommend you to do X and not only a small biased group of people)

**Check out the site and tell me about features you guys want to see**
**ITS 100% FREE AND IM DOING THIS FOR THE LOVE OF THE GAME**

LooksIQ

Community-based face rating. Vote, rank, compete.

looksiq.com

 

[russianlarpcel]

I created the website LOOKSIQ because all existing systems are inherently flawed and viciously circular. Let’s take harmony as an example. Even in studies where ratios are combined with machine learning algorithms they explain only about 60% of the total variance and that’s using a cleaner and more scientific approach than what we typically see here where the scoring system is largely made up (for example 70% harmony= chadlite with no validation).

The modus operandi should not be simply attributing certain ratios to percentiles. Instead one can use ML algorithms like XGBoost to find the relationship between ratios and attractiveness examining both nonlinear and linear relationships. This is done by minimizing a loss function mostly multiclass classification loss by adjusting the weights of each feature in the model. The conclusion is that even near perfectly weighting and combining ratios does not produce an adequate measure of attractiveness.

The alternative I propose is Item Response Theory (IRT) which maps the relationship between the latent variable attractiveness and the likelihood that a person is perceived as more attractive than another person. If enough people vote individual biases cancel out under certain conditions according to the central limit theorem leaving only the true variance being the person’s “true” attractiveness.

Currently I am planning to add features so that users can give advice to others. This advice would then be standardized so that only advice aligned with the consensus of the general population remains ( meaning that most people would recommend you to do X and not only a small biased group of people)

**Check out the site and tell me about features you guys want to see**
**ITS 100% FREE AND IM DOING THIS FOR THE LOVE OF THE GAME**

LooksIQ

Community-based face rating. Vote, rank, compete.

looksiq.com

Why post this in rating

 

[194Cm True MMTN]

I created the website LOOKSIQ because all existing systems are inherently flawed and viciously circular. Let’s take harmony as an example. Even in studies where ratios are combined with machine learning algorithms they explain only about 60% of the total variance and that’s using a cleaner and more scientific approach than what we typically see here where the scoring system is largely made up (for example 70% harmony= chadlite with no validation).

The modus operandi should not be simply attributing certain ratios to percentiles. Instead one can use ML algorithms like XGBoost to find the relationship between ratios and attractiveness examining both nonlinear and linear relationships. This is done by minimizing a loss function mostly multiclass classification loss by adjusting the weights of each feature in the model. The conclusion is that even near perfectly weighting and combining ratios does not produce an adequate measure of attractiveness.

The alternative I propose is Item Response Theory (IRT) which maps the relationship between the latent variable attractiveness and the likelihood that a person is perceived as more attractive than another person. If enough people vote individual biases cancel out under certain conditions according to the central limit theorem leaving only the true variance being the person’s “true” attractiveness.

Currently I am planning to add features so that users can give advice to others. This advice would then be standardized so that only advice aligned with the consensus of the general population remains ( meaning that most people would recommend you to do X and not only a small biased group of people)

**Check out the site and tell me about features you guys want to see**
**ITS 100% FREE AND IM DOING THIS FOR THE LOVE OF THE GAME**

LooksIQ

Community-based face rating. Vote, rank, compete.

looksiq.com

yo wais i know you

 

[wais5psl]

Why post this in ratingWhy post this in rating

Im not that familiar with the structure of the website so if I selected the wrong category I apologize for that lol.

 

[wais5psl]

yo wais i know you

whats your discord?

 

[194Cm True MMTN]

whats your discord?

saturn

 

[194Cm True MMTN]

whats your discord?

i loss my org main

 

[wais5psl]

saturn

Oh nice to meet you here lol.

 

[kdev]

oh wow fancy seeing you here either way add a rater leaderboard

 

[Final Fantasy]

I created the website LOOKSIQ because all existing systems are inherently flawed and viciously circular. Let’s take harmony as an example. Even in studies where ratios are combined with machine learning algorithms they explain only about 60% of the total variance and that’s using a cleaner and more scientific approach than what we typically see here where the scoring system is largely made up (for example 70% harmony= chadlite with no validation).

The modus operandi should not be simply attributing certain ratios to percentiles. Instead one can use ML algorithms like XGBoost to find the relationship between ratios and attractiveness examining both nonlinear and linear relationships. This is done by minimizing a loss function mostly multiclass classification loss by adjusting the weights of each feature in the model. The conclusion is that even near perfectly weighting and combining ratios does not produce an adequate measure of attractiveness.

The alternative I propose is Item Response Theory (IRT) which maps the relationship between the latent variable attractiveness and the likelihood that a person is perceived as more attractive than another person. If enough people vote individual biases cancel out under certain conditions according to the central limit theorem leaving only the true variance being the person’s “true” attractiveness.

Currently I am planning to add features so that users can give advice to others. This advice would then be standardized so that only advice aligned with the consensus of the general population remains ( meaning that most people would recommend you to do X and not only a small biased group of people)

**Check out the site and tell me about features you guys want to see**
**ITS 100% FREE AND IM DOING THIS FOR THE LOVE OF THE GAME**

LooksIQ

Community-based face rating. Vote, rank, compete.

looksiq.com

Can you add a feature to see what most people picked. Im enjoying rating other people but also want to know the % of people who have also voted for these 2 specific photos in the past have voted beforehand.

Like x amount of people voted for left. etc

 

[wais5psl]

Can you add a feature to see what most people picked. Im enjoying rating other people but also want to know the % of people who have also voted for these 2 specific photos in the past have voted beforehand.

Like x amount of people voted for left. etc

We have added the rater alignment score which you can see when clicking on the dashboard. We don’t want to include information after every rate as it will result in bias. If person X won against person Y and you see that the difference was huge you will tend to vote for X again in a different match. That would amplify existing bias and make the bias not normally distributed. If it’s not normally distributed the central limit theorem doesn’t apply anymore and would make the score which you receive less accurate. Hopefully that helps

 

[Wiggle_24]

I created the website LOOKSIQ because all existing systems are inherently flawed and viciously circular. Let’s take harmony as an example. Even in studies where ratios are combined with machine learning algorithms they explain only about 60% of the total variance and that’s using a cleaner and more scientific approach than what we typically see here where the scoring system is largely made up (for example 70% harmony= chadlite with no validation).

The modus operandi should not be simply attributing certain ratios to percentiles. Instead one can use ML algorithms like XGBoost to find the relationship between ratios and attractiveness examining both nonlinear and linear relationships. This is done by minimizing a loss function mostly multiclass classification loss by adjusting the weights of each feature in the model. The conclusion is that even near perfectly weighting and combining ratios does not produce an adequate measure of attractiveness.

The alternative I propose is Item Response Theory (IRT) which maps the relationship between the latent variable attractiveness and the likelihood that a person is perceived as more attractive than another person. If enough people vote individual biases cancel out under certain conditions according to the central limit theorem leaving only the true variance being the person’s “true” attractiveness.

Currently I am planning to add features so that users can give advice to others. This advice would then be standardized so that only advice aligned with the consensus of the general population remains ( meaning that most people would recommend you to do X and not only a small biased group of people)

**Check out the site and tell me about features you guys want to see**
**ITS 100% FREE AND IM DOING THIS FOR THE LOVE OF THE GAME**

LooksIQ

Community-based face rating. Vote, rank, compete.

looksiq.com

Yooo ima look at this site

 

[wais5psl]

Yooo ima look at this site

If you upload your pic be somewhat patient as your score will become accurate within 1-3 days. Additionally consider that the scores for the high range aren’t super accurate yet as only 500 people have uploaded pictures so far.inflation/deflation in the area of +- 2 standard deviations is possible but will be gone in the future.

 

[wais5psl]

Yooo ima look at this site

Also if you like it a post about it would be super helpful. The more people participate the better it becomes.

 

[Wiggle_24]

If you upload your pic be somewhat patient as your score will become accurate within 1-3 days. Additionally consider that the scores for the high range aren’t super accurate yet as only 500 people have uploaded pictures so far.inflation/deflation in the area of +- 2 standard deviations is possible but will be gone in the future.

It won’t lemme upload tho I put in everything

 

[wais5psl]

It won’t lemme upload tho I put in everything

I manually accept/ reject pictures as you can’t accurately automate it

 

[wais5psl]

It won’t lemme upload tho I put in everything

and accepted

 

[Wiggle_24]

M

and accepted

hmm?

 

[slixzn]

I created the website LOOKSIQ because all existing systems are inherently flawed and viciously circular. Let’s take harmony as an example. Even in studies where ratios are combined with machine learning algorithms they explain only about 60% of the total variance and that’s using a cleaner and more scientific approach than what we typically see here where the scoring system is largely made up (for example 70% harmony= chadlite with no validation).

The modus operandi should not be simply attributing certain ratios to percentiles. Instead one can use ML algorithms like XGBoost to find the relationship between ratios and attractiveness examining both nonlinear and linear relationships. This is done by minimizing a loss function mostly multiclass classification loss by adjusting the weights of each feature in the model. The conclusion is that even near perfectly weighting and combining ratios does not produce an adequate measure of attractiveness.

The alternative I propose is Item Response Theory (IRT) which maps the relationship between the latent variable attractiveness and the likelihood that a person is perceived as more attractive than another person. If enough people vote individual biases cancel out under certain conditions according to the central limit theorem leaving only the true variance being the person’s “true” attractiveness.

Currently I am planning to add features so that users can give advice to others. This advice would then be standardized so that only advice aligned with the consensus of the general population remains ( meaning that most people would recommend you to do X and not only a small biased group of people)

**Check out the site and tell me about features you guys want to see**
**ITS 100% FREE AND IM DOING THIS FOR THE LOVE OF THE GAME**

LooksIQ

Community-based face rating. Vote, rank, compete.

looksiq.com

Hi bhai i know you

 

[Wiggle_24]

I created the website LOOKSIQ because all existing systems are inherently flawed and viciously circular. Let’s take harmony as an example. Even in studies where ratios are combined with machine learning algorithms they explain only about 60% of the total variance and that’s using a cleaner and more scientific approach than what we typically see here where the scoring system is largely made up (for example 70% harmony= chadlite with no validation).

The modus operandi should not be simply attributing certain ratios to percentiles. Instead one can use ML algorithms like XGBoost to find the relationship between ratios and attractiveness examining both nonlinear and linear relationships. This is done by minimizing a loss function mostly multiclass classification loss by adjusting the weights of each feature in the model. The conclusion is that even near perfectly weighting and combining ratios does not produce an adequate measure of attractiveness.

The alternative I propose is Item Response Theory (IRT) which maps the relationship between the latent variable attractiveness and the likelihood that a person is perceived as more attractive than another person. If enough people vote individual biases cancel out under certain conditions according to the central limit theorem leaving only the true variance being the person’s “true” attractiveness.

Currently I am planning to add features so that users can give advice to others. This advice would then be standardized so that only advice aligned with the consensus of the general population remains ( meaning that most people would recommend you to do X and not only a small biased group of people)

**Check out the site and tell me about features you guys want to see**
**ITS 100% FREE AND IM DOING THIS FOR THE LOVE OF THE GAME**

LooksIQ

Community-based face rating. Vote, rank, compete.

looksiq.com

Hey bro bro I uploaded this photo and It’s not doin nothin

 

[Pentadecant]

yo how come it doesnt let me upload

 

[wais5psl]

I apologize for the inconvenience. I have exams atm and additionally Im working on the V2 of looksiq which is my priority. I manually evaluate each picture as there is no algorithm which adequate accuracy sadly.