7 High Dimensional Data Visualization Techniques using Psychedelics

Whether it’s clustering, regression, image features, gene sequencing, explainable neural networks or just looking for patterns, there is nothing more difficult than high dimensional data visualization. Even with animation, there’s only so much you can do when the dimensionality starts to climb over four dimensions. Meanwhile our data is only getting more dimensions as sensors gather more and more data.

Without too much labor for that perfect graph, there is a solution. Using natural and synthetic psychedelics, data scientists have been able to begin finding new and innovative ways to visualize data with thousands of dimensions.

1. Healthcare data visualized using LSD

high dimensional data visualization or Healthcare data using LSD

When it comes to patient data, all manor of information is recorded including nutritional information, drug use, genetic information, nearly anything that could be relevant in a diagnosis. There could be hundreds of data points associated with each patient. To solve this issue, one Bioinformatics student in an MD-PhD program decided to experiment with LSD and was successfully able to graph prostate cancer data using the graph above. Apparently researchers were able to study the chart while still under the psychedelic influence and find several new causes behind the deadly disease shortly after they realized the room wasn’t getting smaller like an accordion.

2. Gas Sensor Arrays visualized using Psilocybin

high dimensional data visualization of Gas Sensor Arrays visualized using Psilocybin

Much like the extra sensors, the data did not make sense until the researcher opened their third eye. With 150,000 attributes for each measurement this data set from a gas sensor array while being exposed to turbulent gas mixtures absolutely needed psychedelics for this high dimensional data visualization. Even though it was a time series data set, Psilocybin was required for a spiritual experience to even grasp what was going on with the chemo resistive gas sensors. While an aero engineering student could not think of an easy way to look at his data, they were by visiting their God within.

3. Mississippi River Pollutants Visualized using Peyote

Mississippi River Pollutants visualized using Peyote

Through the entire Mississippi delta and it’s extensive tributaries, the vast amount of industrial chemicals across time, weather, and location makes a data set extremely challenging to chart. One naturalist and environmental engineer studied the data while under the influence of natural grown Peyote. While on a spirit quest, one researcher found that his spirit animal was a river otter, perfect for the data. After hydrating the man meditated and fasted for three months to produce a simple clustering plot in the concise splatter oil painting above. Apparently, someone in the Kansas City is introducing illegal amounts of lead into the river.

4. Driver Expressions Visualized using 25I-NBOMe

Driver Expressions Visualized using 25I-NBOMe

Most would believe that an image database would be the easiest to visualize. That is until you have 606 samples of images of four driver faces to determine facial expressions. Even with dimensionality reduction, the feature space of the images was too much for typical plotting that would make any sense to assess the computer vision algorithms. One computer vision machine learning engineer, determined to win his kaggle competition visualized his clusters feature space after taking a healthy dose of the synthetic hallucinogen 25I-NBOMe. After a nightmarish dream where his wife’s face was transformed into floating three dimensional cubes, he created an equally nightmarish art installation which apparently represents each driver expression with each cube size and placement according to his feature detection scheme. I’m still not sure if it’s working properly but his graph looks like clusters so I guess it’s working.

5. SETI Data visualized using Molly

high dimensional data visualization of SETI Data using Molly

In the search for extra terrestrial life, radio frequencies, signal patterns, nearby star clusters and many other attributes are associated with each piece of data. Because of the huge search space, it’s extremely difficult to make right from left on the data without a snazzy visualization. That’s why alien signals processing engineer David Levinson took some Molly before synthesizing the vast dataset of outer worldly signals. He determined that the fundamental aspect of earth culture was love and by transforming the data into a colorful seductive picture he could spot the signal anomaly if anything did not make him jump for joy. The image above, while adequately summarizing the data in an unfortunately all erotic manner still does not appear to show any signs of life out there and David’s search continues.

6. Hydraulic Test Rig visualized using Mescaline

high dimensional data visualization of Hydraulic Test Rig using Mescaline

In this hydraulic test rig data set, multivariate time series analysis created data with over 43 thousand attributes for each data point. While the LSTM algorithm crunched through the time series data like a champ after months of tweaking, no one knows what it means. With similar affects of LSD and Psilocybin, one QA test engineer was able to distill all of the multi-sensor fault readings into an easy to understand graph above. Writing a Gary’s Mod plugin, the data was transformed into a three dimensional video game level of absurdist body positions and 1970s wallpaper. Other QA engineers familiar with the hydraulic test rig and also under the influence of Mescaline appeared to understand the readings from the test report without needing infinite pages of colored tables and figures. The program manager, however, did not.

7. Cancerous gene expression visualized using DMT

Cancerous gene expression visualized using DMT

In one dataset, patients with BRCA, KIRC, COAD, LUAD, and PRAD type tumors had their RNA gene expressions sequenced into an over 20k attribute data set. With so many dimensions to the data, researchers would need to take a drug which creates an alternative sense of space, and visiting other dimensions and aliens for an adequate high dimensional data visualization. While studying the data set, one Emory University researcher took short 45 minute bursts of DMT psychosis while running tests on various gene deletion and duplication correlations. Determining that each duplicated gene expression was poison during each DMT trip the researchers compiled the correlations into an image mapped plot representing the poisonous nature of our modern capitalist life style. Don’t take the researchers word for it, that’s what the aliens told them.

If you enjoyed this clickbait listicle please like, share, and subscribe with your email, our twitter handle (@JABDE6), or our facebook group here for weekly content.

Published by B McGraw

B McGraw has lived a long and successful professional life as a software developer and researcher. After completing his BS in spaghetti coding at the department of the dark arts at Cranberry Lemon in 2005 he wasted no time in getting a masters in debugging by print statement in 2008 and obtaining his PhD with research in screwing up repos on Github in 2014. That's when he could finally get paid. In 2018 B McGraw finally made the big step of defaulting on his student loans and began advancing his career by adding his name on other people's research papers after finding one grammatical mistake in the Peer Review process.

Leave a Reply

%d bloggers like this: