Ben, Regarding your thoughts on an eCRM application, I can contribute something. I worked for Siebel systems, the leader in CRM applications, for two years...
I think your idea has merit. There are some other areas where I think AI technology could be of even greater use. I will give an example. Siebels product allows for the handling of Service Requests, which is essentially a record kept of when someone calls in with a problem of some type. Things can be recorded like the following: -who called -when they called -the category of the problem -the sub category -the steps taken to resolve the problem -the solution Companies that have been using Siebels product have now been collecting this data (there are Many more items than above, this is just an example) for years. The problem is this: they don't use the data for much of anything useful. Siebel is now selling analytics where they take the data and pump it into a warehouse for data mining. siebel has prebuilt all the analytics that slice and dice the data every way from sunday. this information is useful. a company can learn things like how many service request occur in a certain period of time. what category of SR's occur the most etc. The real problem resides in the solutions database. This was a really weak part of the product. The solution record was comprised of a total of three fields as I recall. One was an identifier, the second what a description of the problem, and the third was the solution written in a paragraph form. This made the thing essentially useless as it was unstructured and it became very hard for a person working on a problem to find the correct solution. Here's the opportunity as I saw it. There is alot of both problems and solutions stored in the Service Request Records and the Solution database, but in the case of the Solution database, it is extremely unstructured. If the knowledge could be extracted from these items, then it could be presented in a way to both the service agents, and also in web self service, that makes finding and extracting a solution alot easier. As I have not studied the problem in detail, I cannot say that this is a nontrivial task. Case studies would need to be created and customers talked to. If you would like to discuss this further, I would gladly talk with you offline. Especially since I have been getting rather bored doing nothing the last two years thanks to Siebel stock options... Peace Kevin [EMAIL PROTECTED] ----- Original Message ----- From: "Ben Goertzel" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Sunday, November 17, 2002 6:15 PM Subject: RE: [agi] Language translation by computer > > > Alex, > > There is a big leap from rule-based NLP to statistical, machine-learning > based NLP. Your group has taken that leap, which is great. Incorporating > feedback from human chatters is a nifty variation on the stats/ML approach, > but still in my view within that approach. > > I think there's another at least equally big leap to be made from the > statistical, machine learning based NLP to NLP that is based on grounding > semantics in nonlinguistic reality. This step opens a big AGI can of worms, > but is very necessary. And I'll be very curious if you guys choose to take > this leap at some point... > > > Using short sentences gives good results, now its time to make the > > sentences longer. More world-knowledge is needed. > > This is exactly the point. I don't think incorporation of world-knowledge > will ever work very successfully if one is using ONLY ungrounded, > expert-system style knowledge. I think some knowledge gained via experience > in a nonlinguistic world is needed. And the integration of this experience > w/ linguistic capability requires a whole different sort of architecture > from a stats/ML based or rule-based NLP or chat system. > > Nevertheless, it sounds like your work is going in a good direction and it's > great that you've found a way to do some interesting science slowly moving > in the direction of AGI, while making money too... > > -- Ben G > > ------- > To unsubscribe, change your address, or temporarily deactivate your subscription, > please go to http://v2.listbox.com/member/ > ------- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/
