1) Some of these studies are lab-based ('animal models') and others are case studies. Note to self: would be worth dividing into type of study.
2) I'm guessing these were all US based. Note to self: classify according to country, and note that these articles are ONLY the English-language ones - how many additional similar studies are in other languages and thus not accessible to English monoglots?
1) There is a lot of information to datamine. I do have access to an AI that can classify datasets (GPT-3), however it is a proprietary service and it costs money to run, the cost of which is proportionate to data size. It does also have bottleneck limits on input.
2) Not all were US based. A few were UK based. One was European. One was Chinese. A handful were in Spanish. The ones with titles in square brackets (although I removed the brackets from the CSV table to avoid issues with alphabetical sorting) appeared to be translations from non-English peer review journals. I omitted the Spanish ones from the video given my audience is primarily American, however I retained the translated ones.
I concur that other countries' datasets are sorely lacking, and you'd need language competent readers/researchers to get that information. A more well resourced individual could even afford to centralise such information, although I wouldn't be the person for it.
It would be useful to have a sense of the dateline of the studies. Is it possible for your bot to sort them by date?
If not, no worries, but it would be interesting to see any clusters and also the start point and end point of the sequence of studies. Obviously, the publication date isn't the same as the date the study was actually conducted but it would be worth knowing, if it's fairly easy to do. But not a big deal.
I am handicapped by being computer-illiterate. I can't do anything with the file you linked to. I've only got an android phone, no laptop or desktop. That's why I need all the abstracts in a word doc or similar, so I can go through and copy and paste the individual blocks of text to sort them into meaningful subsets.
I have previously been very used to working with bibliographies / references. Not easy on a phone, but doable!
If you bear with me, I'll see what data I can extract. You might have to give me maybe a few weeks to get everything in order. I prefer accuracy over speed unless there's a need for urgency.
A .CSV file can be opened in either LibreOffice Calc (which can be installed for free) or Excel if that is of any help, and is compatible with even old versions of software.
In the interim, here is a PDF with the titles and links, sorted by title ascending and with a number count:
It is unlikely I'll be able to pull all data from the studies, but I will try to pull as much information as I can per your request. I do intend to use automation to classify the titles as well, but all of this takes time, both to write an accurate classifier, generate the data and then sort it in a format that is readable.
I am competent to classify the studies into categories, using the abstracts.
(As Jessica Rose says, the key info they had to report but don't want you to notice is often hidden deep in appendices etc so just using the abstracts likely won't pick up any truthbombs, but it will tell us the basics.) I won't do it as detailed as you did on studies 1 - 22, but it would be good to sort them into basic subsets of lab studies, clinical reports etc.
All we need is a watertight overview of this info to shove at the bastards in charge and say, 'You know this as well as we do. Now acknowledge that!' And then, God willing, use it as the basis of a court case and / or a formal investigation of the heads of specific organisations.
So we don't need to get bogged down in too much detail, just be sure it's all watertight. It's not, imo, to convince anyone at the top of harm - they already KNOW the harm - it's just to set it out clearly and unambiguously.
Two things to note:
1) Some of these studies are lab-based ('animal models') and others are case studies. Note to self: would be worth dividing into type of study.
2) I'm guessing these were all US based. Note to self: classify according to country, and note that these articles are ONLY the English-language ones - how many additional similar studies are in other languages and thus not accessible to English monoglots?
1) There is a lot of information to datamine. I do have access to an AI that can classify datasets (GPT-3), however it is a proprietary service and it costs money to run, the cost of which is proportionate to data size. It does also have bottleneck limits on input.
2) Not all were US based. A few were UK based. One was European. One was Chinese. A handful were in Spanish. The ones with titles in square brackets (although I removed the brackets from the CSV table to avoid issues with alphabetical sorting) appeared to be translations from non-English peer review journals. I omitted the Spanish ones from the video given my audience is primarily American, however I retained the translated ones.
I concur that other countries' datasets are sorely lacking, and you'd need language competent readers/researchers to get that information. A more well resourced individual could even afford to centralise such information, although I wouldn't be the person for it.
It would be useful to have a sense of the dateline of the studies. Is it possible for your bot to sort them by date?
If not, no worries, but it would be interesting to see any clusters and also the start point and end point of the sequence of studies. Obviously, the publication date isn't the same as the date the study was actually conducted but it would be worth knowing, if it's fairly easy to do. But not a big deal.
When it comes to data on large scale, nothing is easy to do.
All takes time to program and verify. We don't want any Tesla car crashes in such a mammoth project.
I am handicapped by being computer-illiterate. I can't do anything with the file you linked to. I've only got an android phone, no laptop or desktop. That's why I need all the abstracts in a word doc or similar, so I can go through and copy and paste the individual blocks of text to sort them into meaningful subsets.
I have previously been very used to working with bibliographies / references. Not easy on a phone, but doable!
If you bear with me, I'll see what data I can extract. You might have to give me maybe a few weeks to get everything in order. I prefer accuracy over speed unless there's a need for urgency.
A .CSV file can be opened in either LibreOffice Calc (which can be installed for free) or Excel if that is of any help, and is compatible with even old versions of software.
In the interim, here is a PDF with the titles and links, sorted by title ascending and with a number count:
https://gitlab.com/TheUnderdog/general-research/-/raw/main/COVID-19-Shot-Questions/Part2/755_Studies_PDF.pdf?inline=false
It is unlikely I'll be able to pull all data from the studies, but I will try to pull as much information as I can per your request. I do intend to use automation to classify the titles as well, but all of this takes time, both to write an accurate classifier, generate the data and then sort it in a format that is readable.
I am competent to classify the studies into categories, using the abstracts.
(As Jessica Rose says, the key info they had to report but don't want you to notice is often hidden deep in appendices etc so just using the abstracts likely won't pick up any truthbombs, but it will tell us the basics.) I won't do it as detailed as you did on studies 1 - 22, but it would be good to sort them into basic subsets of lab studies, clinical reports etc.
All we need is a watertight overview of this info to shove at the bastards in charge and say, 'You know this as well as we do. Now acknowledge that!' And then, God willing, use it as the basis of a court case and / or a formal investigation of the heads of specific organisations.
So we don't need to get bogged down in too much detail, just be sure it's all watertight. It's not, imo, to convince anyone at the top of harm - they already KNOW the harm - it's just to set it out clearly and unambiguously.