@misc{New1676,
	author =	{Isaac Newton},
	howpublished =	{Letter to Hooke},
	month =		{5~} # feb,
	year =		{1676},
}

@misc{FIT1998,
	author =	{Dan Eaves and John Hurst},
	title =		{Standards for the Presentation of Written Assignments},
	howpublished =	{School of Computer Science and Software Engineering,
	Faculty of Information Technology, Monash University, Australia;
	On-line resource},
	year =		{1998},
	url =		{http://www.csse.monash.edu.au/\~{}ajh/teaching/resources/studentguide.pdf},
	note =		{(last accessed June 8, 2005)},
}

@book{Lam1994,
	author =	{Leslie Lamport},
	title = 	{\LaTeX: a document preparation system},
	edition =	{Second},
	year =		{1994},
	publisher =	{Addison-Wesley Publishing Company, Inc.},
	address =	{Reading, MA, USA},
}

@inproceedings{AgS1995,
	author =	{Rakesh Agrawal and Ramakrishnan Srikant},
	title =		{Mining Sequential Patterns},
	booktitle =	{Proceedings of the 11th International Conference on Data
	Engineering},
	editor =	{Yu, P. S. and Chen, A. L. P.},
	address =	{Taipei, Taiwan},
	pages =		{3--14},
	month =		mar,	
	year =		{1995},
}

@article{BoL2000,
	author =	{Jos\'{e} Borges and Mark Levene},
	title =		{A Fine Grained Heuristic to Capture Web Navigation Patterns},
	journal =	{ACM SIGKDD Explorations},
	volume =	{2},
	number =	{1},
	pages =		{40--50},
	month =		jun,
	year =		{2000},
}

@book{WeB1998,
	author =	{C. R. Westphal and T. A. Blaxton},
	title =	{Data Mining Solutions: Methods and Tools for Solving Real-World
	Problems},
	year =		{1998},
	publisher =	{John Wiley \& Sons},
	address =	{New York, USA},
}

@misc{HCH2000,
	author =	{John L. Houle and Wanda Cadigan and Sylvain Henry and Anu
	Pinnamaneni and Sonny Lundahl},
	title =		{Database Mining in the Human Genome Initiative},
	howpublished =	{Bio-databases.com white paper, Amita Corporation, 1420
	Blair Place, Suite 500, Ottawa, Ontario, Canada, K1J 9L8},
	year =		{2000},
	url =		{http://www.biodatabases.com/whitepaper01.html},
}

@string{SP = "Signal Processing"}
@string{NC = "Neural Computation"}

@article{Ade1983,
        author =        {F. Ade},
        title =         {Characterisation of textures by ``eigenfilters''},
        journal =       SP,
        volume =        {5},
        pages =         {451--457},
        year =          {1983}
}

@book{AaK1989,
        author =        {E. Aarts and J. Korst},
        title =         {Simulated annealing and {B}oltzmann machines},
        publisher =     {John Wiley and Sons},
        year =          {1989},
        address =       {New York}
}

@article{AGR1996,
        author =        {Joseph J. Atick and Paul A. Griffin and Redlich, A. Norman},
        title =         {The vocabulary of shape: principal shapes for probing
        perception and neural response},
        journal =       NC,
        volume =        {7},
        number =        {1},
        pages =         {1--5},
        month =         feb,
        year =          {1996},
        keywords =      {principal components, shape perception, eigenmode, eigenhead},
        abstract =      {Humans perceive shape rapidly and effortlessly but have
        great difficulties describing what they perceive. This suggests that
        the representation of shape in the brain is abstract and very unlike
        that used in conscious thought. Here we explore the proposal that this
        representation is matched to the statistical properties of objects in
        the environment. From an ensemble of several hundred laser-scanned
        three-dimensional (3D) human heads we extract the principal components
        which provide a compact basis for head shape. We show that, with good
        accuracy, a given head can be represented by linear combination of a
        few dozen primary shapes just as colours can be synthesized by
        combining the three principal colours. We suggest new perceptual
        adaptation experiments for testing the brain's shape representation
        system. The principal head shapes can also be used to probe response
        properties of `face-cells' in the inferior temporal cortex.}
}


@misc{WiS1994,
        author =        {Peter Williams and Thorsten Schnier},
        title =         {The {H}arvard Family of Bibliography Styles},
        howpublished =  {\LaTeX2e package documentation},
        month =         jun # {~21},
        year =          {1994},
        url =           {http://www.ctan.org/tex-archive/macros/latex/contrib/supported/harvard/},
        note =          {(last accessed July 24, 2002)},
}

@inproceedings{DeV1998,
        vgproject =     {cbir},
        author =        {Jeremy S. {De~Bonet} and Paul Viola},
        title =         {Texture Recognition Using a Non-parametric Multi-Scale
        Statistical Model},
        crossref =      {CVPR98},
        year =          {1998},
        url =           {http://www.ai.mit.edu/\~{}jsd/research/publications/1998/DeBonet-CVPR98.pdf},
        abstract =      {We describe a technique for using the joint occurrence of
        local features at multiple resolutions to measure the similarity
        between texture images. Though superficially similar to a number of
        ``Gabor'' style techniques, which recognize textures through the
        extraction of multi-scale feature vectors, our approach is derived from
        an accurate generative model of texture, which is explicitly
        multi-scale and non-parametric. The resulting recognition procedure is
        similarly non-parametric, and can model complex non-homogeneous
        textures. We report results on publicly available texture databases. In
        addition, experiments indicate that this approach may have sufficient
        discrimination power to perform target detection in synthetic aperture
        radar images (SAR).}
}

@proceedings{CVPR98,
        title = {Proceedings of the 1998 IEEE Conference on Computer Vision and Pattern Recognition (CVPR'98)},
        booktitle =     {Proceedings of the 1998 IEEE Conference on Computer Vision and Pattern Recognition (CVPR'98)},
        key =           {CVPR98},
        month =         jun,
        year =          {1998},
        address =       {Santa Barbara, California, USA}
}`

@misc{SNS2002,
	author =	{David McG. Squire and Fred Quentin Nurke and Karen {Sparck Jones}},
	title =		{Something about {XML}},
	year =		{2002},
}

@misc{SNS2002a,
	author =	{Squire, David McG. and Nurke, Fred Quentin and Sparck Jones, Karen},
	title =		{Something about {XML}},
	year =		{2002},
}
	
@misc{SNS2002b,
	author =	{David McG. Squire and Fred Quentin Nurke and Karen Sparck Jones},
	title =		{Something about XML},
	year =		{2002},
}

@inproceedings{DGV2000,
	author =	{Anjali Dhond and Amar Gupta and Sanjeev Vadhavkar},
	title =		{Data mining techniques for optimizing inventories for
	electronic commerce},
	booktitle =	{Proceedings of the Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
	address =	{Boston, Massachusetts, USA},
	month =		{20--23~} # aug,
	year =		{2000},
	pages =		{480--486},	
	url =		{http://doi.acm.org/10.1145/347090.347188},
	keywords =	{Inventory Optimization, Temporal Data Mining, Data
	Massaging},
	abstract =	{As part of their strategy for incorporating electronic
	commerce capabilities, many organizations are involved in the
	development of information systems that will establish effective
	linkages with their suppliers, customers, and other channel partners
	involved in transportation, distribution, warehousing and maintenance
	activities. These linkages have given birth to comprehensive data
	warehouses that integrate operational data with supplier, customer,
	channel partners and market information. Data mining techniques can now
	provide the technological leap needed to structure and prioritize
	information from these data warehouses to address specific end-user
	problems. Emerging data mining techniques permit the semi-automatic
	discovery of patterns, associations, changes, anomalies, rules, and
	statistically significant structures and events in data. Very
	significant business benefits have been attained through the
	integration of data mining techniques with current information systems
	aiding electronic commerce. This paper explains key data mining
	principles that can play a pivotal role in an electronic commerce
	environment. The paper also highlights two case studies in which neural
	network-based data mining techniques were used for inventory
	optimization. The results from the data mining prototype in a large
	medical distribution company provided the rationale for the strategy to
	reduce the total level of inventory by 50\% (from a billion dollars to
	half a billion dollars) in the particular organization, while
	maintaining the same level of probability that a particular customer s
	demand will be satisfied. The second case study highlights the use of
	neural network based data mining techniques for forecasting hot metal
	temperatures in a steel mill blast furnace.},
}

@article{Fre2000,
	author =	{Alex A. Freitas},
	title =		{Understanding the Crucial Differences Between
	Classification and Discovery of Association Rules---A Position Paper},
	journal =	{SIGKDD Explorations},
	volume =	{2},
	number =	{1},
	pages =		{65--69},
	month =		jun,
	year =		{2000},
	url =		{http://www.acm.org/sigs/sigkdd/explorations/issue2-1/contents.htm\#Freitas},
	keywords =	{Classification, association rules, induction, prediction},
	abstract =	{The goal of this position paper is to contribute to a
	clear understanding of the profound differences between the
	association-rule discovery and the classification task.  We argue that
	the classification task can be considered an ill-defined,
	non-deterministic task, which is unavoidable given the fact that it
	involved prediction; while the standard association task can be
	considered a well-defined, deterministic, relatively simple task, which
	does \emph{not} involve prediction in the same sense as the
	classification task does.},
}

@article{The1999,
	author =	{Kurt Thearling},
	title =		{Data Mining and {CRM}: Zeroing in on Your Best Customers},
	journal =	{DM Direct},
	month =		{20~} # jun,
	year =		{1999},
	note =		{(on-line newsletter)},
	url =		{http://www.dmreview.com/master.cfm?NavID=198\&EdID=1744},
	abstract =	{To be successful, database marketers must first identify
	market segments containing customers or prospects with high profit
	potential and, second, build and execute campaigns that favorably
	impact the behavior of these individuals.

	The first task, identifying the market segments, requires significant
	data about the prospective customers and their buying behaviors. In
	theory, the more data the better. In practice, however, massive data
	stores often impede marketers who struggle to sift through the minutiae
	to find the nuggets of valuable information\ldots},
}

@techreport{ABS1999,
	author =	{Rakesh Agrawal and Roberto Bayardo and Ramakrishnan
	Srikant},
	title =		{Athena: Mining-based Interactive Management of Text
	Databases},
	
	type =		{IBM Research Report},
	number =	{RJ10153},
	institution =	{IBM Almaden Research Center},
	address =	{650 Harry Road, K55/B1 San Jose, CA 95120, USA},
	month =		jul,
	year =		{1999},
	url =		{http://www.almaden.ibm.com/cs/people/srikant/papers/edbt00\_rj.pdf},
	abstract =	{We describe Athena: a system for creating, exploiting, and
	maintaining a hierarchy of textual documents through interactive
	mining-based operations. Requirements of any such system include speed
	and minimal end-user effort. Athena satisfies these requirements
	through linear-time classification and clustering engines which are
	applied interactively to speed the development of accurate models.

	Naive Bayes classifiers are recognized to be among the best for
	classifying text.  We show that our specialization of the Naive Bayes
	classifier is considerably more accurate (7 to 29\% absolute increase
	in accuracy) than a standard implementation. Our enhancements include
	using Lidstone's law of succession instead of Laplace's law,
	under-weighting long documents, and over-weighting author and subject.

	We also present a new interactive clustering algorithm, C-Evolve, for
	topic discovery.  C-Evolve first finds highly accurate cluster digests
	(partial clusters), gets user feedback to merge and correct these
	digests, and then uses the classification algorithm to complete the
	partitioning of the data. By allowing this interactivity in the
	clustering process, C-Evolve achieves considerably higher clustering
	accuracy (10 to 20\% absolute increase in our experiments) than the
	popular K-Means and agglomerative clustering methods.},
}

@inproceedings{ABS2000,
	author =	{Rakesh Agrawal and Roberto Bayardo and Ramakrishnan
	Srikant},
	title =		{Athena: Mining-based Interactive Management of Text
	Databases},
	booktitle =	{Proceedings of the Seventh International Conference on
	Extending Database Technology (EDBT00)},
	address =	{Konstanz, Germany},
	month =		{20--23~} # mar,
	year =		{2000},
	pages =		{365--379},	
	url =		{http://www.almaden.ibm.com/software/quest/Publications/papers/edbt00.pdf},
	abstract =	{We describe Athena: a system for creating, exploiting, and
	maintaining a hierarchy of textual documents through interactive
	mining-based operations. Requirements of any such system include speed
	and minimal end-user effort. Athena satisfies these requirements
	through linear-time classification and clustering engines which are
	applied interactively to speed the development of accurate models.

	Naive Bayes classifiers are recognized to be among the best for
	classifying text.  We show that our specialization of the Naive Bayes
	classifier is considerably more accurate (7 to 29\% absolute increase
	in accuracy) than a standard implementation. Our enhancements include
	using Lidstone's law of succession instead of Laplace's law,
	under-weighting long documents, and over-weighting author and subject.

	We also present a new interactive clustering algorithm, C-Evolve, for
	topic discovery.  C-Evolve first finds highly accurate cluster digests
	(partial clusters), gets user feedback to merge and correct these
	digests, and then uses the classification algorithm to complete the
	partitioning of the data. By allowing this interactivity in the
	clustering process, C-Evolve achieves considerably higher clustering
	accuracy (10 to 20\% absolute increase in our experiments) than the
	popular K-Means and agglomerative clustering methods.},
}

@mastersthesis{Ves1997,
	author =	{Juha Vesanto},
	title =		{Data Mining Techniques Based on the Self-Organizing Map},
	month =		{26~} # may,
	year =		{1997},
	school =	{Helsinki University of Technology},
	address =	{Laboratory of Computer and Information Science, P.O. Box
	5400, FIN-02015 HUT, Finland},
	url =		{http://www.cis.hut.fi/projects/ide/publications/html/mastersJV97/},
	keywords =	{neural network, Self-Organizing Map, data mining,
	knowledge discovery, pulp and paper industry},
	abstract =	{Data mining is a part of a larger area of recent research
	in artificial intelligence and information management: knowledge
	discovery in databases (KDD). The purpose of KDD is to find new
	knowledge from databases in which the dimension, complexity or the
	amount of data has so far been prohibitively large for human
	observation alone. Data mining refers to the exploratory phase of
	knowledge discovery.

	The Self-Organizing Map (SOM) is one of the most popular neural network
	models. The SOM quantizes the data space formed by the training data
	and simultaniously performs a topology-preserving projecting of the
	data space on a regular two-dimensional grid. The SOM also has
	excellent visualization capabilities including techniques to give an
	informative picture of the data space, and techniques to compare data
	vectors or whole data sets with each other. The SOM can also be used
	for clustering, classification and modeling. The versatile properties
	of the SOM make it a valuable tool in data mining and knowledge
	discovery.

	As part of this work a SOM-based data mining tool was implemented. The
	methods and tools presented in the work were used to analyze the pulp
	and paper industry worldwide and the Scandinavian industry in more
	detail with encouraging results. The analysis of technological data
	resulted in 20 major types of pulp and paper mills. Regarding
	Scandinavian industry a hierarchical structure of SOMs was used to
	combine technological, environmental and economical data.

	The work has been done in the Laboratory of Computer and Information
	Science at the Helsinki University of Technology as part of the
	corporate project Entire in the technology program ``Adaptive and
	Intelligent Systems Applications''. The project was financed by Jaakko
	P\"{o}yry Consulting and the Technology Development center of Finland
	(TEKES).},
}

@misc{Dal1999,
	author =	{Patrick W. Daly},
	title =		{Natural Sciences Citations and References (Author--Year
	and Numerical Schemes)},
	howpublished =	{\LaTeX2e package documentation},
	month =		may # {~28},
	year =		{1999},
	url =		{http://www.ctan.org/tex-archive/macros/latex/contrib/natbib/},
	note =		{(last accessed February 7, 2004)},
}

@article{Car2005,
	author =	{Scott Carlson},
	title =		{Scholars Note `Decay' of Citations to Online References},
	journal =	{The Chronicle of Higher Education},
	month =		{18~} # mar,
	year =		{2005},
	volume =	{51},
	number =	{28},
	pages =		{A.30},
	abstract =	{Michael Bugeja says that when he got his doctorate in
	English, he studied the difference between ``fair'' and ``foul'' copies of
	Shakespeare's plays -- a foul copy being rife with inaccuracies.

	``That's because the medium of printing was unstable back then,'' says
	Mr.  Bugeja, a professor of journalism and communication at Iowa State
	University.

	Now that the Internet is the new unstable publishing medium, he and a
	colleague have studied how Web links stop working, or ``decay,'' as
	those sites change addresses or shut down. They focused on links used
	by scholars in footnotes that cite Web materials.

	After analyzing more than 1,126 citations that make reference to Web
	addresses, taken from online versions of five prestigious
	communication-studies journals, 373 of the links, or 33 percent, were
	found to be dead. Of the 753 of the links that worked, only 424 pointed
	to information pertinent to the citation.

	Mr. Bugeja and Daniela Dimitrova, an assistant professor of
	communication at Iowa State, looked at footnotes from 2000 to 2003 in
	Human Communication Research, the Journal of Broadcasting \& Electronic
	Media, the Journal of Communication, Journalism \& Mass Communication
	Quarterly, and New Media \& Society.

	``The erosion of footnotes,'' Mr. Bugeja says, ``might put us back to a
	curious situation, wondering whether we have a fair copy of a journal
	article or a foul copy of a journal article.''

	In some journals, the decay rate was particularly high. For example, of
	the 265 citations in New Media \& Society articles that included links,
	167 did not work.

	Steve Jones, a professor of communication at the University of Illinois
	at Chicago who is an editor of ``New Media \& Society,'' called the
	decay of online citations "a real issue" that the journal has begun to
	examine.

	He wonders whether copyright law might someday allow scholars to copy
	and archive online articles that they used as sources. But he says such
	a solution is ``pie in the sky.''

	Anthony T. Grafton, a professor of history at Princeton University who
	has written a book about footnotes, has read a draft of the study and
	agrees that citation decay is ``a real problem.''

	``I'm looking at a world in which documentation and verification melt
	into air,'' he says. He sees this problem growing, as today's students
	rely more on online sources. ``My students come to college less and
	less able to negotiate a book landscape and more and more adept at
	negotiating the Web.''

	Mr. Bugeja and Ms. Dimitrova are preparing their findings for
	publication and are coming up with a list of recommendations to stop
	the decay of online citations. Their findings and recommendations will
	be presented at the International Communication Association conference
	in May.},
}