Results 1 to 9 of 9

Thread: The Future of Research

  1. #1

    The Future of Research

    Hello, everyone I have an idea that could potentially lead to cures and treatments for diseases that as of yet have no satisfactory treatment or cure. My idea, is the idea of generating new ideas, here is how it works:

    We can use the Grid concept to process the following:

    Make a giant computerized database of nearly all the most important medical research papers, second each paper will be analyzed for word patterns, word frequency, word usage, grammar, etc. A linguistic model of each paper will be made and categorized.

    Then combining words, word pairs, sentences, at random or following certain parameters will generate a new abstract according to some criteria as set by the linguistic model. Now to check that the abstract might contain new ideas, it will checked against the database of existing papers, a linguistic metric will be used to gauge how "new" or different the abstract is, and to make sure the idea is new, an internet search will be performed to determine how many websites or papers contain the same phrases, words, word usage, etc. Another check to make sure the idea is new will take random sentences from the paper and checked using google. First the entire sentence will be searched for then search for the sentence leave one word from the search phrase, then leave off two words from the search phrase, then three words and so on. Those abstracts with the lowest average hits amongst all the searches and the highest average linguistic metric distance will be collected and viewed.

    I take it for granted that many of papers may not make sense given that they are put together at random, but the random jumble of words and sentences might trigger a new idea in the hopefully competent person reading the abstract.

    If one is looking for a new idea closer to an existing idea, all one needs to do is collect all the papers on that idea, and change the wording and word usage in the paper very very slightly, so the idea more or less makes sense yet might contain new possibilities. One can make a list of words not often used and insert them into the paper.

    Another idea is to make a database of the papers only in one particular part of one field, then analyze the word usage and word pattern and then make a statistical model to help predict the words and ideas that might be used in papers in the near future. One can employ a genetic algorithm to modify the statistical algorithm to make the predicted
    papers correlate closely with papers in the near future. You are trying to predict the field in a systematic manner.

    Now lets say you are looking for a new treatment or cure, so what you can do is generate all the abstracts with the words that might appear in the paper that will lead to this new treatment or cure, amongst all those abstracts there may exist the abstract that will lead to the new treatment or cure.

    Of course, there are many details to be worked out, but I believe that if implemented correctly it will lead to new treatments and cures that we would have never thought of.

    John G.

  2. #2
    Are you serious?? Are you HopeforSCI????
    T6 complete (or so I think), SCI since September 21, 2003

  3. #3
    Senior Member roshni's Avatar
    Join Date
    Nov 2005
    Location
    Old Bridge, New Jersey, USA
    Posts
    2,496
    FSC729,

    So in your paradigm, shifting phrases around might yield the cure?

  4. #4
    LoL.....Your SCI has you delusional already.

  5. #5
    Senior Member Quadcessible's Avatar
    Join Date
    Aug 2001
    Location
    Florida
    Posts
    150

    Talking

    This guy probably has the Bible Code figured out too

  6. #6
    Senior Member
    Join Date
    Apr 2006
    Location
    New York City
    Posts
    251
    Actually It sounds like the way a chess computer works. The computer goes through a history of moves and positions that have been reached before. It eventually makes a desicion based on what either worked previously or did not work. If it cannot find the position because none exists, then it proceeds to something similar and moves accordingly. It's basically a process of elimination and of course the need to control territory and eliminate danger.

    There was a time not so long ago when human beings were of the opinion that a computer would never be able to beat a champion player because of the human's propensity for surprise and in general all things human. Of course the humans were proved wrong.

    There was also a position that at one time was thought to be unwinnable. However the computer in this case through sheer might, managed to once again prove humans wrong.

  7. #7
    Senior Member Quadcessible's Avatar
    Join Date
    Aug 2001
    Location
    Florida
    Posts
    150
    "Actually It sounds like the way a chess computer works. The computer goes through a history of moves and positions that have been reached before. It eventually makes a desicion based on what either worked previously or did not work. If it cannot find the position because none exists, then it proceeds to something similar and moves accordingly. It's basically a process of elimination and of course the need to control territory and eliminate danger.

    There was a time not so long ago when human beings were of the opinion that a computer would never be able to beat a champion player because of the human's propensity for surprise and in general all things human. Of course the humans were proved wrong.

    There was also a position that at one time was thought to be unwinnable. However the computer in this case through sheer might, managed to once again prove humans wrong."

    You have a point, but consider this. There are 64 squares on a checker board of which each opponent has 16 pieces, which have the ability to move in certain configurations. Through out these configurations they have a mathamatical method by which a computer can calculate all moves on the board and quickly find the best solution because all the movement of the parameters are known due to the limits imposed by the playing broard and the pieces (Closed-Loop). At this point in SCI research there is much that is known and much to discover. Much of what is known is being studied, replicated and combined to learn more (Open-Loop). The computer is only as good as the information that it has stored in its memory, or basically what is known, it can't figure out what to do with what it doesn't know. It can help in calculating the best chances or statistics of success in selecting certain therapys which might help SCI but the bottom line is the scientists combining the best of what they know and begin appling them to optimize the best they can give us today. We would all like the one magic therapy today, but its just not there yet, but maybe its time to start appling some of what they know to begin improving our lives, sure there's some risk, to a point a risk many of us would consider taking.

  8. #8

    Like flipping a coin!!

    Its a small world and things don’t need to be so complicated or so easy
    just simple.

    If we get the scientist to work together, at least here in the US,
    Have healthy competition against other continents,
    Set a time goal and have a world convention to match all the research and than last, find the cheapest well equipped country and start monkey trials, than we will be closer to a real "cure".
    Glider

  9. #9
    They truly do come out at night.

Similar Threads

  1. NIH priorities from 2003 to 2007
    By Wise Young in forum Cure
    Replies: 19
    Last Post: 04-15-2006, 06:50 AM
  2. Norway: The stem cell research potential and politics
    By Leif in forum Funding, Legislation, & Advocacy
    Replies: 4
    Last Post: 03-31-2006, 04:42 AM
  3. Remarkable turnaround in media reporting concerning stem cell research in the past we
    By Wise Young in forum Funding, Legislation, & Advocacy
    Replies: 3
    Last Post: 03-01-2006, 05:41 AM
  4. Replies: 3
    Last Post: 08-31-2004, 01:14 PM
  5. Replies: 0
    Last Post: 08-16-2002, 11:50 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •