[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Subscribe]
Re: DM: Datamining Definition...and Machine Learning Definition.From: Franklin Wayne Poley Date: Wed, 22 Mar 2000 22:51:11 -0800 Thank you Santro. This seems to be what the University of East Anglia is saying more succintly on its web site, http://www.sys.uea.ac.uk/PGStudy/mscke.htm re the MSc in knowledge extraction which it defines as "extracting knowledge from larger databases". I used the word "text" below because I thought the word "database" might be too broad. The whole universe is a data base in a sense...mountainous for sure. If we go with such a broad term then data mining/knowledge extraction becomes synonymous with machine learning does it not? Would an Msc in machine learning then be the same as an Msc in data extraction? The other web site given was http://www.cs.bris.ac.uk/Teaching/MachineLearning . Elsewhere I have come up with 11 categories of machine learning and I will forward them to the list later. "Learning from text" would be one category. I think most people would want machine learning to stand for a much broader field than data mining/knowledge extraction. What I am trying to get at with "learning from text" is that this is one of the most common and incredibly complex ways in which humans learn. Imagine any text book. It contains as you note below, words, numbers, symbols and pictures. The same information can be presented by the teacher in a classroom ; the text is turned into an oral presentation. If I could speed through text books without assistance from teachers I would quickly become the most learned person in the world. It is quite amazing that students extract the "noteworthy" or "valuable" material from all of this information. I think phrases like data mining and knowledge extraction may communicate better with cs specialists but "learning from text" may communicate better outside this field. It sounds very much to me at this stage that we would mean the same thing. After a lot of reading texts and going to lectures where the texts are explained, elaborated on etc. the medical student extracts the noteworthy data on how to do medical diagnoses. Now what would it do to medical training if the student could use a computer/robot to extract the same "hidden knowledge" from the "larger databases" of the medical texts, lectures etc.? Before even trying to answer that question I want to make sure we have pretty good agreement on the definition. Also would you agree that data mining is one kind of machine learning? Here is another: maze learning. I think machines have this one mastered do they not? The algorithms for maze learning are quite simple compared to those for data mining are they not? I wonder then which category of machine learning we would place at one end of that simple-complex continuum and which we would place at the other? FWP. -----Original Message----- From: Santrokofi <santrokofi@yahoo.com> To: datamine-l@nautilus-sys.com <datamine-l@nautilus-sys.com> Date: Wednesday, March 22, 2000 9:28 PM Subject: Re: DM: Datamining Definition. >Hi Franklin, > >I hope this definition helps: > >"Data mining explores mountainous databases, using >automated approaches, to reveal meaningful patterns. > The databases may contain numbers, words, graphs, or >pictures. Data from these different sources can be >pooled into data warehouses. Data mining algorithms >can then examine numerous multidimensional data >relationships concurrently, highlighting those that >are noteworthy." > >Santro Kofi > >--- Franklin Wayne Poley <culturex@vcn.bc.ca> wrote: > > > > Could datamining be fairly defined as "extracting > > data of value from > > text"? (With "text" broadly defined to include > > words, sentences, pictures, > > symbols etc? > > FWP. > > >http://users.uniserve.com/~culturex/Machine-Psychology.htm
|
MHonArc
2.2.0