This is the time of the year when Technology Review publishes its forecasts about ten emerging technologies which will change our world some day. This year's batch includes Bayesian machine learning, RNA interference or microfluidic optical fibers. But last year's list included injectable tissue engineering or nanoimprint lithography, which didn't really change the world in 2003. So read this list with a grain of salt.
Let's start with the introduction.
With new technologies constantly being invented in universities and companies across the globe, guessing which ones will transform computing, medicine, communication, and our energy infrastructure is always a challenge. Nonetheless, Technology Review’s editors are willing to bet that the 10 emerging technologies highlighted in this special package will affect our lives and work in revolutionary ways -- whether next year or next decade. For each, we’ve identified a researcher whose ideas and efforts both epitomize and reinvent his or her field.
Here is the full list.
Universal Translation, with Yuqing Gao, from IBM
Synthetic Biology, with Ron Weiss, from Princeton University
Nanowires, with Peidong Yang of the University of California, Berkeley
Bayesian Machine Learning, with Daphne Koller, from Stanford University
T-Rays, with Don Arnone, from Toshiba’s research labs in Cambridge, England
Distributed Storage, with Hari Balakrishnan, from the MIT
RNA Interference, with Thomas Tuschl, formerly from the Max Planck Institute for Biophysical Chemistry in Germany, and now at Rockefeller University in New York City
Power Grid Control, with Christian Rehtanz, from Switzerland-based engineering giant ABB
Microfluidic Optical Fibers, with John Rogers, from the University of Illinois
Personal Genomics, with David Cox, chief scientific officer of Perlegen Sciences in Mountain View, CA
Here is the last paragraph of the article about nanowires.
Difficult tasks remain, such as making electrical connections between the minuscule wires and the other components of any system. Still, Peidong Yang of the University of California, Berkeley, estimates there are now at least 100 research groups worldwide devoting significant time to overcoming such obstacles, and commercial development efforts have already begun. Last year, Intel, which is working with Lieber, revealed that nanowires are part of its long-term chip planning. Smaller firms such as Nanosys and QuMat Technologies, a startup now renting space at Lund University in Sweden, are betting that nanowires will be essential components of the products they hope to sell one day, from sensors for drug discovery and medical diagnosis to flat-panel displays and superefficient lighting.
And here is a short excerpt about Bayesian statistics.
Programs that employ Bayesian techniques are already hitting the market: Microsoft Outlook 2003, for instance, includes Bayesian office assistants. English firm Agena has created Bayesian software that recommends TV shows to satellite and cable subscribers based on their viewing habits; Agena hopes to deploy the technology internationally. "These things sound far out," says Microsoft researcher Eric Horvitz, who is a leading proponent of probabilistic methods. "But we are creating usable tools now that you’ll see in the next wave of software."
Thursday, February 26, 2009
A robotic Cyberknife to fight cancer
The Cyberknife is not a real knife. This is a robot radiotherapy machine which works with great accuracy during treatment, thanks to its robotic arm which moves around a patient when he breathes. According to BBC News, the first Cyberknife will be operational in February 2009 in London, UK. But other machines have been installed in more than 15 countries, and have permitted to treat 50,000 patients in the first semester of 2008. And the Cyberknife is more efficient than conventional radiotherapy devices. The current systems require twenty or more short sessions with low-dose radiation. On the contrary, and because it's extremely precise, a Cyberknife can deliver powerful radiation in just three sessions. ...
As you can see above, "the CyberKnife System uses image guidance software to track and continually adjust treatment for any patient or tumor movement. This sets it far ahead of other similar treatments. It allows patients to breathe normally and relax comfortably during treatment." And it "uses pencil beams of radiation which can be directed at any part of the body from any direction via a robotic arm;" (Credit: various Accuray pages). Here is a link to a larger version of thie picture above.
The Cyberknife is a product of Accuray, who has deployed a corporate CyberKnife website and many other local sites in various locations. Speaking about locations, here is a link to a page where you can check if you live in a place not too far from a medical institution using such a system.
Here is an excerpt from the BBC News article. "At first sight the Cyberknife looks like one of those robots used in the TV car commercials. It is a compact linear accelerator mounted on a robot arm. The cyberknife works by delivering multiple beams of high dose radiation from a wide variety of angles using a robotic arm. X-ray cameras monitor the patient's breathing and re-position the radiotherapy beam in order to minimise damage to healthy tissue. This accuracy enables tumours to be treated that are in difficult or dangerous to treat positions, such as near the spinal cord."
Of course, such a treatment is expensive. "Treatment will cost between £20,000 and £25,000."
Now, let's look at the new CyberKnife Centre in London to discover what is CyberKnife and how it works."The vast array of different angles/trajectories from which pencil beams of radiation converge upon the tumour lead to an extremely high cumulative dose of radiation therapy at the convergence point (the target/tumour) and yet a very fast 'fall-off' of dose at the periphery of the carefully mapped target. The surrounding normal tissues/organs only receive a small fraction of the high central dose of therapy."
You'll also find explanations about why the Cyberknife is about to replace surgery -- at least in some cases. "The treatment is so accurate that it's now possible to treat tumours previously thought to be inoperable. Although the results of treatment do not always show immediately, in most cases the procedure will initially stop the growth of tumours before gradually reducing their size. As there is no open surgery, the complications normally associated with an operation are eliminated, as is the need for a long recovery time. This makes treatment suitable for those who are not well enough to cope with the side-effects of surgery and most patients leave the clinic the same day as their treatment."
As you can see above, "the CyberKnife System uses image guidance software to track and continually adjust treatment for any patient or tumor movement. This sets it far ahead of other similar treatments. It allows patients to breathe normally and relax comfortably during treatment." And it "uses pencil beams of radiation which can be directed at any part of the body from any direction via a robotic arm;" (Credit: various Accuray pages). Here is a link to a larger version of thie picture above.
The Cyberknife is a product of Accuray, who has deployed a corporate CyberKnife website and many other local sites in various locations. Speaking about locations, here is a link to a page where you can check if you live in a place not too far from a medical institution using such a system.
Here is an excerpt from the BBC News article. "At first sight the Cyberknife looks like one of those robots used in the TV car commercials. It is a compact linear accelerator mounted on a robot arm. The cyberknife works by delivering multiple beams of high dose radiation from a wide variety of angles using a robotic arm. X-ray cameras monitor the patient's breathing and re-position the radiotherapy beam in order to minimise damage to healthy tissue. This accuracy enables tumours to be treated that are in difficult or dangerous to treat positions, such as near the spinal cord."
Of course, such a treatment is expensive. "Treatment will cost between £20,000 and £25,000."
Now, let's look at the new CyberKnife Centre in London to discover what is CyberKnife and how it works."The vast array of different angles/trajectories from which pencil beams of radiation converge upon the tumour lead to an extremely high cumulative dose of radiation therapy at the convergence point (the target/tumour) and yet a very fast 'fall-off' of dose at the periphery of the carefully mapped target. The surrounding normal tissues/organs only receive a small fraction of the high central dose of therapy."
You'll also find explanations about why the Cyberknife is about to replace surgery -- at least in some cases. "The treatment is so accurate that it's now possible to treat tumours previously thought to be inoperable. Although the results of treatment do not always show immediately, in most cases the procedure will initially stop the growth of tumours before gradually reducing their size. As there is no open surgery, the complications normally associated with an operation are eliminated, as is the need for a long recovery time. This makes treatment suitable for those who are not well enough to cope with the side-effects of surgery and most patients leave the clinic the same day as their treatment."
Old Apple Hard Drive Becomes New Atomic Mirror
Before going further, what is an "atomic mirror"? As its name implies, it reflects atoms instead of light. In this article, NewsFactor Network tells us that Cal Tech researchers have fabricated such a mirror using an old Apple disk drive.
"An atom mirror is a device that reflects impinging atoms in an analogous manner to the way a regular optical mirror reflects an incoming light beam," said Cal Tech quantum-optics researcher Benjamin Lev. "The difference between the optical mirror and the atom mirror is that ... to reflect photons one only needs a suitable metallic surface, [but] to reflect atoms one needs to create some sort of repulsive force for the atoms as they near the surface."
The Cal Tech research team "fabricated a magnetic mirror by etching a common hard drive, and used this mirror to reflect a cold cloud of cesium atoms," wrote lead researcher Hideo Mabuchi.
The common hard drive, Mabuchi explained, has several features that make it the perfect raw material for atomic mirror makers -- a large, flat, magnetic surface; smooth contours; and rigid construction.
Here is how the etched hard drive looks like.
But what can we do with such a thing?
The "atomic mirror" ultimately may help engineers create atomic lasers, ushering in new telecommunications technologies based on atoms rather than photons, and atomic -- not electronic -- signals.
Atomic optics may find application in another burgeoning tech field that remains in early infancy -- quantum-computer science. "One exciting prospect is to use the atom mirror, combined with electric fields, to perform quantum logic gates necessary for building a quantum computer."
"An atom mirror is a device that reflects impinging atoms in an analogous manner to the way a regular optical mirror reflects an incoming light beam," said Cal Tech quantum-optics researcher Benjamin Lev. "The difference between the optical mirror and the atom mirror is that ... to reflect photons one only needs a suitable metallic surface, [but] to reflect atoms one needs to create some sort of repulsive force for the atoms as they near the surface."
The Cal Tech research team "fabricated a magnetic mirror by etching a common hard drive, and used this mirror to reflect a cold cloud of cesium atoms," wrote lead researcher Hideo Mabuchi.
The common hard drive, Mabuchi explained, has several features that make it the perfect raw material for atomic mirror makers -- a large, flat, magnetic surface; smooth contours; and rigid construction.
Here is how the etched hard drive looks like.
But what can we do with such a thing?
The "atomic mirror" ultimately may help engineers create atomic lasers, ushering in new telecommunications technologies based on atoms rather than photons, and atomic -- not electronic -- signals.
Atomic optics may find application in another burgeoning tech field that remains in early infancy -- quantum-computer science. "One exciting prospect is to use the atom mirror, combined with electric fields, to perform quantum logic gates necessary for building a quantum computer."
Nanocomputing: Simple Optoelectronic Devices Perform Logic Operations
Researchers at the Georgia Institute of Technology have demonstrated a new type of nanometer-scale optoelectronic (which combine light and electronics technologies) device that can perform addition and other complex logic operations.
These quantum devices are based on arrays of individual electroluminescent silver nanoclusters.
"In effect, we are demonstrating optoelectronic transistor behavior," said Robert Dickson, a professor in Georgia Tech's School of Chemistry and Biochemistry. "Instead of measuring current output as in standard electronic transistors, we measure electroluminescent output for a given voltage input. Our devices act in a way that is analogous to a transistor with light as the output instead of electrical current."
Each cluster contains between two and eight silver atoms, and emits light when electrically excited by a specific voltage. Operating the device requires a pair of separate electrical pulses, the second of which generates electroluminescence when specific nanoclusters are activated according to voltage level.
Individual clusters can operate as logic gates with AND, OR, NOT, and XOR functions via the application of different pulses, while more complex operations can be performed by increasing the number of clusters.
"By using this complicated on-off behavior and the discrete energy levels of different molecules, we can get complicated behavior in a relatively simple device," said Dickson.
Increasing the number of clusters operating together could permit formation of large optoelectronic arrays able to perform complex operations. As long as each cluster could be separated enough to be resolved by a camera, arrays could contain thousands of clusters.
Dickson doesn't expect the new optoelectronic devices to replace traditional semiconductor-based computers for ordinary tasks. Instead, they might be used for complex and highly specialized computations that are difficult for traditional computing systems.
He also hopes that the breakthrough will inspire other researchers to reconsider nanometer-scale computing.
This research is funded by the National Science Foundation and will be reported in the March 18 issue of the journal Proceedings of the National Academy of Sciences.
These quantum devices are based on arrays of individual electroluminescent silver nanoclusters.
"In effect, we are demonstrating optoelectronic transistor behavior," said Robert Dickson, a professor in Georgia Tech's School of Chemistry and Biochemistry. "Instead of measuring current output as in standard electronic transistors, we measure electroluminescent output for a given voltage input. Our devices act in a way that is analogous to a transistor with light as the output instead of electrical current."
Each cluster contains between two and eight silver atoms, and emits light when electrically excited by a specific voltage. Operating the device requires a pair of separate electrical pulses, the second of which generates electroluminescence when specific nanoclusters are activated according to voltage level.
Individual clusters can operate as logic gates with AND, OR, NOT, and XOR functions via the application of different pulses, while more complex operations can be performed by increasing the number of clusters.
"By using this complicated on-off behavior and the discrete energy levels of different molecules, we can get complicated behavior in a relatively simple device," said Dickson.
Increasing the number of clusters operating together could permit formation of large optoelectronic arrays able to perform complex operations. As long as each cluster could be separated enough to be resolved by a camera, arrays could contain thousands of clusters.
Dickson doesn't expect the new optoelectronic devices to replace traditional semiconductor-based computers for ordinary tasks. Instead, they might be used for complex and highly specialized computations that are difficult for traditional computing systems.
He also hopes that the breakthrough will inspire other researchers to reconsider nanometer-scale computing.
This research is funded by the National Science Foundation and will be reported in the March 18 issue of the journal Proceedings of the National Academy of Sciences.
Microsoft's Palladium -- or TCP/MS
named Palladium, which got lots of attention from the press last week. The Palladium feature should appear around 2006 at the same time as the next generation of Windows, code-named Longhorn.
So why this concern about a product which will not appear before four years (or more, we're dealing with Microsoft)? Simply because it will invade your privacy. And because it will cost you lots of money (again, we are talking about Microsoft.)
As wrote Reuters on July 2, "Instead of storing sensitive information such as passwords on software, Palladium will also aim to protect information at the hardware level."
But let's use some words from a really talented columnist, Robert X. Cringely, to give you a clearer view of what will represent Palladium for you.
Last August, I wrote of a rumor that Microsoft wanted to replace TCP/IP with a proprietary protocol -- a protocol owned by Microsoft -- that it would tout as being more secure. Actually, the new protocol would likely be TCP/IP with some of the reserved fields used as pointers to proprietary extensions. I called it TCP/MS in the column.
This week, Microsoft announced Palladium through an exclusive story in Newsweek written by Steven Levy, who ought to have known better. Palladium is the code name for a Microsoft project to make all Internet communication safer by essentially pasting a digital certificate on every application, message, byte, and machine on the Net, then encrypting the data EVEN INSIDE YOUR COMPUTER PROCESSOR. Palladium compatible hardware (presumably chipsets and motherboards) will come from both AMD and Intel, and the software will, of course, come from Microsoft. That software is what I had dubbed TCP/MS.
The point of all this is simple. It may actually make the Internet somewhat safer. But the real purpose of this stuff, I fear, is to take technology owned by nobody (TCP/IP) and replace it with technology owned by Redmond. That's taking the Internet and turning it into MSN. Oh, and we'll all have to buy new computers.
This is diabolical. If Microsoft is successful, Palladium will give Bill Gates a piece of every transaction of any type while at the same time marginalizing the work of any competitor who doesn't choose to be Palladium-compliant. So much for Linux and Open Source, but it goes even further than that. So much for Apple and the Macintosh. It's a militarized network architecture only Dick Cheney could love.
So why this concern about a product which will not appear before four years (or more, we're dealing with Microsoft)? Simply because it will invade your privacy. And because it will cost you lots of money (again, we are talking about Microsoft.)
As wrote Reuters on July 2, "Instead of storing sensitive information such as passwords on software, Palladium will also aim to protect information at the hardware level."
But let's use some words from a really talented columnist, Robert X. Cringely, to give you a clearer view of what will represent Palladium for you.
Last August, I wrote of a rumor that Microsoft wanted to replace TCP/IP with a proprietary protocol -- a protocol owned by Microsoft -- that it would tout as being more secure. Actually, the new protocol would likely be TCP/IP with some of the reserved fields used as pointers to proprietary extensions. I called it TCP/MS in the column.
This week, Microsoft announced Palladium through an exclusive story in Newsweek written by Steven Levy, who ought to have known better. Palladium is the code name for a Microsoft project to make all Internet communication safer by essentially pasting a digital certificate on every application, message, byte, and machine on the Net, then encrypting the data EVEN INSIDE YOUR COMPUTER PROCESSOR. Palladium compatible hardware (presumably chipsets and motherboards) will come from both AMD and Intel, and the software will, of course, come from Microsoft. That software is what I had dubbed TCP/MS.
The point of all this is simple. It may actually make the Internet somewhat safer. But the real purpose of this stuff, I fear, is to take technology owned by nobody (TCP/IP) and replace it with technology owned by Redmond. That's taking the Internet and turning it into MSN. Oh, and we'll all have to buy new computers.
This is diabolical. If Microsoft is successful, Palladium will give Bill Gates a piece of every transaction of any type while at the same time marginalizing the work of any competitor who doesn't choose to be Palladium-compliant. So much for Linux and Open Source, but it goes even further than that. So much for Apple and the Macintosh. It's a militarized network architecture only Dick Cheney could love.
Rack 'n Roll: Why you should take a Mac user to lunch
Apple Computer, Inc. introduced the Xserve system about two months ago. I don't know if the product will be successful, but the specifications are impressive.
You can pack 2 G3 processors with 2 gigabytes of memory and 480 gigabytes of disk storage in 1U unit. Or you can have 84 processors with almost 20 terabytes of storage in one big rack, for a peak performance of 630 gigaflops per second.
And if it wasn't enough, Apple has a great motto. Here it is.
All the details about this server are here.
After this -- rather admiring -- introduction, let's see what LinuxWorld has to say about the Unix market and the Xserve.
Although the actual numbers are bit fuzzy, it seems that Sun leads the Unix market in terms of the number of users served, IBM leads in Unix related revenues, and Apple, not Dell or HP, sells the most Unix boxes. Those machines run the MacOS X layer on top of Darwin, an open source BSD variant with a MACH kernel.
Like Linux, the underlying Unix for MacOS X is an open source production and, again like Linux, it has all the traditional Unix virtues including high reliability, network compatibility, efficient resource use, and access to a wide variety of lower cost, cutting-edge tools and applications.
Apple should ship almost 4 million Unix desktops this year, and each one of them represents a new opportunity for open source ideas to take root and for products like OpenOffice.org to find users. Equally importantly, each time a Mac moves into an office environment it gets harder to maintain the fiction that homogeneous (meaning all Windows) systems are cheaper or easier to run.
Paul Murphy compares prices for similar systems from Sun, Dell and Apple boxes with different OSes. Guess what: Apple is the cheapest.
He also compares hardware and software prices of a desktop system running a Microsoft operating system today and twenty years ago.
Here are the numbers for 1981.
Hardware: $2,959
Microsoft OS: $39.95
Now, let's come back to 2002.
Hardware: $450
Microsoft OS: $199
In other words, the hardware price decreased by 85 percent while the operating system from Microsoft increased by 500 percent.
You can pack 2 G3 processors with 2 gigabytes of memory and 480 gigabytes of disk storage in 1U unit. Or you can have 84 processors with almost 20 terabytes of storage in one big rack, for a peak performance of 630 gigaflops per second.
And if it wasn't enough, Apple has a great motto. Here it is.
All the details about this server are here.
After this -- rather admiring -- introduction, let's see what LinuxWorld has to say about the Unix market and the Xserve.
Although the actual numbers are bit fuzzy, it seems that Sun leads the Unix market in terms of the number of users served, IBM leads in Unix related revenues, and Apple, not Dell or HP, sells the most Unix boxes. Those machines run the MacOS X layer on top of Darwin, an open source BSD variant with a MACH kernel.
Like Linux, the underlying Unix for MacOS X is an open source production and, again like Linux, it has all the traditional Unix virtues including high reliability, network compatibility, efficient resource use, and access to a wide variety of lower cost, cutting-edge tools and applications.
Apple should ship almost 4 million Unix desktops this year, and each one of them represents a new opportunity for open source ideas to take root and for products like OpenOffice.org to find users. Equally importantly, each time a Mac moves into an office environment it gets harder to maintain the fiction that homogeneous (meaning all Windows) systems are cheaper or easier to run.
Paul Murphy compares prices for similar systems from Sun, Dell and Apple boxes with different OSes. Guess what: Apple is the cheapest.
He also compares hardware and software prices of a desktop system running a Microsoft operating system today and twenty years ago.
Here are the numbers for 1981.
Hardware: $2,959
Microsoft OS: $39.95
Now, let's come back to 2002.
Hardware: $450
Microsoft OS: $199
In other words, the hardware price decreased by 85 percent while the operating system from Microsoft increased by 500 percent.
A 'Smart' Email Software Organizes Your Tasks
You probably receive dozens of emails every day about various aspects of your business or personal life. And because your email program doesn't understand the relationship between messages, except for the occasional thread, you have to manage your activities by looking through lists of emails. But now, two computer scientists from University College Dublin (UCD) and IBM have developed the Active Email Manager (AEM) and have even filed patents for a 'smart' email program. Their prototype can make the difference between work-related tasks -- and assign them to a workflow -- and personal email. This software could be integrated in commercial products from IBM within two years. Read more...
Here are some details about the project.
A University College Dublin (UCD) scientist has filed a patent application for a new technology that he believes can turn email into a much more effective business tool. US-born Dr Nicholas Kushmerick, a senior lecturer in the Department of Computer Science at UCD, has developed the technology over the past year during his part-time position as visiting scientist on IBM’s Centre for Advanced Studies (CAS) initiative.
Kushmerick developed the technology, known as Active Email Manager (AEM), in concert with New York-based IBM researcher Tessa Lau. Together they developed a machine-learning algorithm that automatically keeps track of tasks and associated emails, in order to build up a work flow for each task.
"The vision is that rather than come in and download all your emails, you could just call up your to do list and manage your activities," Kushmerick explains.
Now, the two researchers have developed a prototype of the software and are busy testing it. And IBM wants to use the technology in some of its future products.
The technology is currently being appraised by two separate research groups within IBM, with the aim of turning into a commercial product. One of these is the Massachusetts-based product development team that develops IBM’s suite of collaboration software, Lotus Workplace. "There are some pretty intensive discussions going on now to see if we can get enough attention and convince them that our idea is feasible and that they would put it into their product pipeline," says Kushmerick.
The research work has been presented at the 2005 International Conference on Intelligent User Interfaces (IUI 2005) which has been held on January 9-12, 2005, in San Diego, California. You can find the abstract of the paper called "Automated Email Activity Management: An Unsupervised Learning Approach" in the 2005 Conference Program.
Many structured activities are managed by email. For instance, a consumer purchasing an item from an e-commerce vendor may receive a message confirming the order, a warning of a delay, and then a shipment notification. Existing email clients do not understand this structure, forcing users to manage their activities by sifting through lists of messages. As a first step to developing email applications that provide high-level support for structured activities, we consider the problem of automatically learning an activity's structure. We formalize activities as finite-state automata, where states correspond to the status of the process, and transitions represent messages sent between participants. We propose several unsupervised machine learning algorithms in this context, and evaluate them on a collection of e-commerce email.
Please note that this work received a Honorable Mention for Outstanding Paper Award at IUI 2005.
For more information, here is a link to the full version of this paper (PDF format, 8 pages, 234 KB), available from Kushmerick's website
Here are some details about the project.
A University College Dublin (UCD) scientist has filed a patent application for a new technology that he believes can turn email into a much more effective business tool. US-born Dr Nicholas Kushmerick, a senior lecturer in the Department of Computer Science at UCD, has developed the technology over the past year during his part-time position as visiting scientist on IBM’s Centre for Advanced Studies (CAS) initiative.
Kushmerick developed the technology, known as Active Email Manager (AEM), in concert with New York-based IBM researcher Tessa Lau. Together they developed a machine-learning algorithm that automatically keeps track of tasks and associated emails, in order to build up a work flow for each task.
"The vision is that rather than come in and download all your emails, you could just call up your to do list and manage your activities," Kushmerick explains.
Now, the two researchers have developed a prototype of the software and are busy testing it. And IBM wants to use the technology in some of its future products.
The technology is currently being appraised by two separate research groups within IBM, with the aim of turning into a commercial product. One of these is the Massachusetts-based product development team that develops IBM’s suite of collaboration software, Lotus Workplace. "There are some pretty intensive discussions going on now to see if we can get enough attention and convince them that our idea is feasible and that they would put it into their product pipeline," says Kushmerick.
The research work has been presented at the 2005 International Conference on Intelligent User Interfaces (IUI 2005) which has been held on January 9-12, 2005, in San Diego, California. You can find the abstract of the paper called "Automated Email Activity Management: An Unsupervised Learning Approach" in the 2005 Conference Program.
Many structured activities are managed by email. For instance, a consumer purchasing an item from an e-commerce vendor may receive a message confirming the order, a warning of a delay, and then a shipment notification. Existing email clients do not understand this structure, forcing users to manage their activities by sifting through lists of messages. As a first step to developing email applications that provide high-level support for structured activities, we consider the problem of automatically learning an activity's structure. We formalize activities as finite-state automata, where states correspond to the status of the process, and transitions represent messages sent between participants. We propose several unsupervised machine learning algorithms in this context, and evaluate them on a collection of e-commerce email.
Please note that this work received a Honorable Mention for Outstanding Paper Award at IUI 2005.
For more information, here is a link to the full version of this paper (PDF format, 8 pages, 234 KB), available from Kushmerick's website
Subscribe to:
Posts (Atom)
