In one of the previous comments in the past, dear "Cvetka" wrote about a topic that I never considered, creating the impression that the idea she wrote was my own thought/approach. What she wrote was: "According to Ahmet's logic, only Turkish linguists can judge the VM and claim or disclaim whether the language is actually Old Turkic."
You are not allowed to view links.
Register or
Login to view.
However, there is a fact that I have been repeating here for a long time. I have written many times that there is no need to be a linguist or know Turkish to understand or verify the evidence we present regarding VM. Now, in addition to this, I have shown that "even artificial intelligence can help you understand that certain results have been achieved on these matters or provide clues.
From time to time here, I have mentioned the structural evidence I presented in my articles, which are seen only in Turkish and VM texts. I have also stated that these have been seen in other Turkish manuscripts in history, by explaining the manuscript names and referring to academic studies.
At the same time, I pointed out many linguistic pieces of evidence using linguistic methods and finding words on dictionary pages. I mentioned that more than a hundred sentences, more than a hundred drawing-word matches, many full pages, and over 1000 words have been read in the VM texts. I also talked about our articles sent to international peer-reviewed/scientific-board symposiums related to the Turkish language and history, which have been read and published at an academic level in their journals and books.
Later, I indicated what the evidence we occasionally presented to you showed to people researching this subject and asked some questions. I never received a clear and equally scientific response to my very specific questions based on very clear evidence that I asked many times. Instead of answering my numerous questions, many of you preferred to remain silent here. Now, when artificial intelligence started forming sentences implying that Turkish is the most likely result, you collectively entered a race to criticize artificial intelligence.
What exactly is your aim?
Are there any experts among you in artificial intelligence software, artificial intelligence mathematical approaches, and fuzzy logic?
As an electrical engineer with a degree in engineering science, I am someone who knows how calculators and computers, as well as electrical automation with open and closed circuit switching between zero and one, work. If you are going to criticize artificial intelligence, please say consistent things beyond general and personal opinions.
I have already mentioned in my previous comments that artificial intelligence currently has many errors and shortcomings. But at the same time, machines can more quickly touch on searching, finding, sorting, comparing, calculating, and accessing millions of written sources, thousands of dictionaries, and hundreds of linguistic articles in different languages compared to humans.
As with everything, nowadays, everyone has an opinion about the working principle of artificial intelligence. I know that having an opinion without having proper knowledge is a human condition. Machines can evaluate the information in data pools in certain ways and criteria, and the ways and speeds of their functioning are changing day by day.
The project of getting humans to make machines think is actually a much older academic project than most of you know.
Mathematician Ordinaryus Prof. Dr. Cahit Arf (1910 - 1997) carried out studies on the thinking of machines in 1958. One of his articles, "Can a Machine Think and How Can It Think?" (Makine Düşünebilir mi, Nasıl Düşünebilir?) His scientific article titled was published at Atatürk University in the 1958 -1959 academic year. In addition to making observations about the possibility of machines having some features that can be considered indicators of a person's thinking, Arf presents us with machine design examples that can convince us that they can think. According to him, machines; It can be designed with mental abilities such as using language, calculating, thinking based on analogy and elimination, and performing logical and analytical operations, and there are similarities between the way the human brain works and the way machines work. However, Arf sees the main difference between humans and machines in the difficulty of imparting the aesthetic consciousness that humans have to machines. In this context, the study tried to point out the similarities and differences between humans and machines with the arguments put forward by Arf in his article, and Arf's thinking machine designs were discussed. His studies on the solvability of synthetic geometry problems have brought the theorem called "Hasse-Arf Theorem" to the science of mathematics, as well as studies mentioned in the literature such as the Arf constant and Arf rings, regarding the invariants that occur in the classification of quadratic forms of objects. He coined terms such as "Arf Constant" and "Arf Closings".
You are not allowed to view links.
Register or
Login to view.
&
You are not allowed to view links.
Register or
Login to view.
&
You are not allowed to view links.
Register or
Login to view.
According to Azerbaijani scientist Lutfi Zadeh (1921 - 2017); the degree of truth of each proposition takes continuous values between true and false (or between zero and one). "Zadə's logic" can see real life more realistically. This topic is one of Lutfi Zadeh's 6 important theories. His theory of fuzzy logic has been a guide for scientists in terms of robotics and computer thinking. This theory gave a new expression to the concept of binary set, which is the basis of mathematics: fuzzy set. The inclusion of a fuzzy dimension in science allows for more adequate consideration of the uncertainty of the processes taking place in nature and society. Lutfi Zade's work, known as the Z transformation in science, is a scientific theory that lays the foundation for the creation of discrete and digital control, information, and communication systems. His famous state space, control and observation theories of dynamic systems form the basis of modern management science. He introduced the concepts of switching with complex logic, known as fuzzy logic/intelligence, in addition to open and closed circuit logic, to the known zero and one switching system of calculators and computers. This logic intellectually opened the door to consider the possibilities between zero and one when switching in the electronic space.
I asked the following question to the artificial intelligence below, and its answer to my question can be read in the images below, with its clear, consistent, and realistic presence.
QUESTION to AI:
As artificial intelligence, do you have the ability to compare information in the knowledge pool and distinguish what is logical and more likely true from what is illogical or more likely false? Therefore, do you have the ability to distinguish between concrete evidence and abstract claims in your measurements, comparisons, and analyses, and add new information or a new result to your own data pool as a result of these? Or to put it briefly: Do you have the ability to draw conclusions or learn from your own experiences?
REPLY:
[
attachment=8647]
[
attachment=8648]
[
attachment=8649]
Now, leaving aside some inconsistent and controversial personal speculations and denunciations on AI,
let's summarize the current actual AI capacity at the scientific level from GPT-AI's own answer on this subject:
* Artificial intelligence (AI) has some capabilities to compare information, assess its consistency, and identify patterns.
* AI can identify patterns in large datasets and check for internal consistency.
* AI can differentiate between concrete, specific data (e.g., numerical data, factual statements) and abstract concepts (e.g., philosophical ideas) to some extent.
* AI systems, especially those based on machine learning, can "learn" from data by adjusting their models based on new inputs. This process, called training, allows AI to improve its performance on specific tasks over time.
* AI learning is fundamentally different from human learning. It is limited to the scope and quality of the training data and does not involve conscious experience or understanding. Additionally, AI systems require significant amounts of labeled data and computational resources to learn effectively.
* AI can draw conclusions based on statistical correlations and predefined algorithms. For instance, AI can predict trends, classify data, and make decisions within defined parameters.
* The conclusions drawn by AI are heavily dependent on the input data and existing algorithms.
* Truth often requires context, which AI may not fully grasp. Detecting misinformation involves understanding nuances that go beyond textual analysis.
* AI has impressive capabilities in data analysis, pattern recognition, and learning from structured data,
* AI "learning" is based on data and algorithms rather than conscious experience, and its ability to draw meaningful conclusions is constrained by the limitations of its training and programming.
In conclusion:
As a group that does not want to answer the clear and specific questions I have asked about numerous details of the Voynich manuscript, I thank you for sharing your negative opinions on the extent to which artificial intelligence can do the job correctly. When AI models like GPT-3.5 and GPT-4max state that "there is a higher probability that the VM content may be in Turkish, but not the higher probability that other languages such as Indo-European languages, Semitic languages, and Latin (including other languages claimed for the VM) can be considered with a smaller probability," you should reconsider knowing that the current capacity of AI includes the ability to compare the consistency of academic papers written on this topic. Science progresses through the evaluation of evidence, and researchers who cannot properly evaluate evidence cannot achieve accurate results in the scientific field. Moreover, in the scientific field, "whether or not individuals like the concrete results explained based on evidence holds no value." What matters is the rationality, consistency, and accuracy of the scientific outcome. In science, nationalist or preferred viewpoints should hold no value. Don't get me wrong, I am not writing this for you, but to reiterate a fact.
Here, I have shown numerous different pieces of evidence that the VM texts have Turkish content. I have also expressed the accuracy of this result with mathematical probabilities. Furthermore, I have compared all academic papers that claim the VM contains a natural language with artificial intelligence and shared its opinion with you. The AI examined the topic and also highlighted the probability that the VM texts may align more with Turkish rather than other languages previously suggested.
With all these pieces of evidence and results in front of us; is there a researcher among you who can say, "
I think the same as AI. & the evidence you presented shows that there is a higher probability that the VM content could be in Turkish"?
Thanks,
Note 1: I asked the artificial intelligence its thoughts about a word written by Mevlânâ Celâleddîn-i Rûmî (1207-1273). The AI answered this as follows:
[
attachment=8650]
[
attachment=8651]
Note 2: When you ask artificial intelligence a question that is poorly constructed or not aligned with its operational logic, the answers you receive may not be scientific or realistic. However, if you educate yourself on how AI logic works and how questions should be posed, AI can make scientific and realistic inferences. You and thousands of others have made numerous claims about the Voynich manuscript, with some of them being academic, and others (individuals) may be potentially numbering in the thousands in general. The most common information in the internet's knowledge pool and written articles (with millions of results) is that the Voynich manuscript cannot be read. When this widespread misconception and information is expected to be the outcome of artificial intelligence, the machine's explanation that the Voynich manuscript is more likely to be in Turkish is solely because it has read and evaluated the consistent and rational evidence presented in our three articles.
Note 3: Please, no one reading my comments on this page should be offended by the succinct quote Mevlana said in the 13th century. However, this approach concerns how it conditions the subconscious mind and whether it has the potential to transcend the boundaries it sets for itself. If the boundaries, patterns, and ideas you believe to be true cannot be changed by evidence proving them wrong, this situation is not related to the person presenting the evidence. So, there is no need to blame me for what I have written. If you are honest with yourself and not dogmatic (which I hope you are not), you will make an effort to check whether the presented evidence is correct. If you do not want to make such an effort, it shows that you are happy with your current knowledge. When it was proven that the world was not actually flat like a plate, this knowledge did not change quickly either. Today, we know that there are still people who believe that the world is flat. My struggle with VM readings stems from the fact that I have difficulty understanding & translating some words from some of the 600-year-old European languages in VM into Turkish sentences, and I seek help on this matter. However, if you still do not understand that the alphabet transcription you have does not work, you obviously will not be able to provide the intellectual work and help I expect. But this situation does not prevent me from stating the evidence too.