| |
| ||||||
IAPA - Institute of Analytic Professionals of AustraliaThis is a discussion on IAPA - Institute of Analytic Professionals of Australia within the Analytics forums, part of the Subject Matter Expertise category; I'm a member of IAPA and find great value in it. They hold regular meetings in various cities around Australia. The website is here . I recommend joining! This thread ... |
 |
| | LinkBack | Thread Tools | Search this Thread | Display Modes |
13th November 2008, 12:48 PM
| #1 |
| Member |  IAPA - Institute of Analytic Professionals of Australia I'm a member of IAPA and find great value in it. They hold regular meetings in various cities around Australia. The website is here. I recommend joining! This thread is to talk about IAPA, post copies of newsletters, anything. |
|      |
|
13th November 2008, 12:56 PM
| #2 |
| Administrator |  IAPA Alert - Issue 33 (12 November 2008) If you're having trouble viewing this email, click here to read the alert online The Newsletter Archive is now online , click here to view  12 November 2008  IAPA Alert - Issue 33 In today's Newsletter
Message from the Chair I'm sure I write on behalf of all members in passing on our gratitude to our departing Chair, Inna Kolyshkina and Secretary, Warwick Graco. Both have served with passion and commitment to steer the IAPA from its inception to what it is today, an organisation that is nearing 1,000 members representing professionals in many disciplines of analytics across multiple industries across Australia. The IAPA's mission is to unite, inform, support and promote business analytics professionals in Australia. We provide information sources, a virtual community, a networking hub and a professional identity. We promote the benefits of analytics in modern business. We emphasise the increasingly strategic role played by data analysis in the business arena. We are committed to developing recognition of analytics professionals in the wider business community and establishing standards of practice, excellence, accreditation and education. The Board of IAPA have identified a number of initiatives to deliver further value to our members, sponsors and partners throughout 2009. These initiatives include: Benchmarking - with access to nearly 1,000 local Analytics Professionals we have the opportunity to collate and define best practice across numerous analytic disciplines; by participating in this exercise you will have the opportunity to leverage this unique asset to improve the use of analytics within your organisation. Mentoring - we intend to call on our extensive and experienced network to provide mentoring to our newer members and people transitioning to this growing discipline. Alignment - we will be developing and leveraging relationships with relevant professional bodies to bring our members benefits of an extended network into related disciplines, exposure to relevant forums and events and to further grow the IAPA brand and membership listening to you - IAPA ceases to remain relevant without your input; we encourage you to provide feedback via the forthcoming Member Survey to provide us with your ideas as to how you'd like to see IAPA evolve. Website - we will be re-launching an improved online presence (www.iapa.org.au) to enable more member collaboration and participation. As a Not-for-Profit organisation we are wholly reliant on our Board, Chapter Heads, Fellows and Members to devote valuable time outside of their everyday roles to fuel our momentum. With Eugene Dubossarsky and myself taking on Board responsibilities there are two high profile openings for the Head of NSW Chapter and the Victorian Chapter respectively. These roles are at the heart of the Analytics communities in Sydney and Melbourne, providing you with an opportunity to expand your professional network and provide guidance and leadership to the local IAPA activities. If you are motivated and passionate about analytics and want to make a difference to this profession, please contact me at docampbell@deloitte.com.au to express your interest. On a final note, with these challenging economic times we, as analytic professionals, are provided with an even louder voice to ensure our organisations are doing more with the finite (and in some cases, constricting) resources available to them; what better environment to prove the value of applying analytics than now? I look forward to working with and hearing from you in the coming year, Doug Campbell. Editorial The attention of most nations in the last two months has been on the economic meltdown in the USA and Europe and the amount of toxic debt carried by financial institutions. There has much finger pointing with senior management of banks and other financial houses and credit agencies coming in for much criticism because of their high salaries and very generous retirement packages. While it is easy to finger point, it is suggested that what is required is not an attack on executives' remuneration packages as this is very likely to be counterproductive. It is important to encourage responsible risk taking and to reward successful performance if investors and for that matter taxpayers are to reap the benefits of economic growth and prosperity. What needs attention are the incentives that encourage corporate bad behaviour. At present there is an asymmetry between risk and reward with executive remuneration. If executives take large risks with shareholders' funds and they win they are usually handsomely rewarded. In contrast if they lose they can retire on lucrative packages while shareholders are usually left carrying the losses and taxpayers have to bail out irresponsible financial institutions to keep the wheels of the economy turning. It is these perverse incentives of rewarding rather than punishing reckless decisions that need to be changed. Another issue that has to be addressed is the strategic drivers of meltdowns such as that experienced in the housing sector in the USA. This meltdown has threatened to bring down the rest of the economy in the USA and in turn the world. Mark Latham in the Australian Financial Review of Thursday 30 October 2008, P66 pointed out that governments have reinforced the dream of universal home ownership and provided various incentives for families to put themselves into mortgage debt beyond their means. This has encouraged financial institutions to loan monies to people who could not afford the repayments based on the expectation that they could recoup their losses by selling the defaulters' homes that were rising in market value. This led to foreclosures and significant losses when the housing bubble burst and the prices of the houses dropped dramatically in value. Governments too have to exercise responsibility by providing programs that achieve important social goals without seriously distorting the economy. Finally there is the question of regulation. It is a truism that too little regulation is an open invitation to exploitation, fraud and abuse while too much regulation tends to stifle initiative and undermine investment and economic growth. It is a challenge for governments to achieve the right balance with regulation. What is important with regulating markets is having early warning systems and appropriate risk-management strategies in place. The former is required to detect economic tsunamis that can do much harm while the latter provide mitigation measures to contain the damage that can be done. It also pays to have a policeman on the beat in the markets to stamp out minor crimes before they become major crimes. Historical Analysis A technique that is very important with the current financial meltdown is historical analysis. It is an imperative that society learns from previous historical examples such as the 1929 Depression to avoid repeating history. There are two items below which highlight issues with this approach to analysis. Dr Ed Lewis a risk management and systems planning lecturer and consultant from the School of Information Technology and Electrical Engineering at the University of NSW at the Australian Defence Force Academy is an advocate of historical analysis. He is married to a history teacher so he has little choice. He makes the following points at his website. Historical analysis is an important technique in risk management For example, in disaster risk management, historical analysis of previous incidents helps to determine levels of risk and, as described by Shi (2006), it " ... examines historical events to gain insight into potential future risks". ... but it has its dangers. Shi (2006) points out that many of the events of interest are rare and with negative consequences. It might not be of help in identifying risks factors for conditions that have never occurred before, as in new technological changes or business practices. It might be misleading if the historical conditions are not applicable in the present or the future. Historical analysis is more than history Historical analysis is not just about recording what happened. It is about why it happened and whether it will again. Historical analysis is not just history. It is also analysis. So, it uses the philosophy of causation and the statistical techniques of correlation; with systems theory to understand the effect of feedback. As Santayana has been repeatedly quoted as saying: Those who cannot remember the past are condemned to repeat it, but noting that he meant "A man's memory may almost become the art of continually varying and misrepresenting his past, according to his interests in the present.". In other words, we have the Theory of Planned Behaviour or Perceptual Control Theory at work. Historical analysis should draw upon an understanding also of human nature, of the quirks of individual decision-making and of group behaviour. Nonetheless, historical analysis helps in the Resources-Risk-Required-Remedy thinking process. It helps to:
Ed's website is below In another source on this issue it is noted that military historians cannot manipulate independent variables and conduct randomized experiments to establish causal relationships, as do other scientists. History presents observational data, making it impossible for the historian to state that "x causes y." The historian's goal is to document a relationship or discover an association between two or more variables in the targeted set of cases, without establishing causality. Usually a victory or defeat in a particular battle depends on a host of factors, and many of those factors are often interrelated. For example, can the historian use case studies in which a non-swarming army defeats a swarming army and attribute that victory to any one variable such as concentration of mass? It is very difficult to know what other variables such as command, logistics, morale, or even blind luck play in determining the outcome. When Crassus's Roman legions were cut to pieces by the swarming Parthians in 53 B.C., was the Roman defeat due to incompetent leadership or the asymmetric balance between highly mobile horse archers and slow-moving infantrymen? Any single variable--such as technology--cannot be viewed in isolation. It follows that, in studying historical cases, it is often helpful to combine observation with an evolving theory--to identify the likely key factors and relationships among factors early in order to be better able to "see" and interpret historical cases with some sophistication. The purpose is to evolve good, structured insights, even if rigorous and precise conclusions cannot be drawn from the historical cases alone. Forecasting This subject was covered in the previous newsletter. Attached are a number of very useful guides on forecasting provided by Dr J. Scott Armstrong from The Wharton School, University of Pennsylvania, Philadelphia. His website Forecastingprinciples.com was created in 1997 and he is the Director. Dr Kesten C. Green of the Business and Economic Forecasting Unit at Monash University Australia, joined him in 2006 as co-director. Document 1 (Word) Document 2 (.pdf) Document 3 (.pdf) Data Fusion Dr Simson L. Garfinkel is an associate professor in computer science at the Naval Postgraduate School in Monterey California. He has written an insightful article on the limitations of data fusion in the Scientific American September 2008 edition. He points out that this is both a difficult and challenging process and the hardest step is making sense of the fused information. The article can be obtained from the link below Notices AUSDM 2008 A timely reminder that AUSDM is on in Adelaide on 27-28 November 2008 Australasian Data Mining Conference Stamford Grand, Glenelg, Adelaide 27-28 November 2008 We would like to invite you to attend this year's Australasian Data Mining Conference (AusDM'08) in Adelaide. AusDM has established itself as the premier Australasian meeting for both practitioners and researchers in data mining. We are delighted that for this year's conference we have three exciting keynote speakers, a tutorial on privacy-preserving data sharing and data mining, a PhD consortium, and 24 peer-reviewed papers selected from 55 submitted papers. The preliminary program for AusDM'08 can be found at: http://ausdm08.togaware.com/preliminaryProgram.html KEYNOTES * Prof Geoff Webb, Monash University Title: "MultiStrategy Ensemble Learning, Ensembles of Bayesian Classifiers, and the Problem of False Discoveries" * Prof David Powers, Flinders University of South Australia Title: "Minors as Miners: Modelling and Evaluating Ontological and Linguistic Learning" * Dr Richard Price, Defence Science and Technology Organisation Title: "Volume, Velocity and Variety - Key Challenges for Mining Large Volumes of Multimedia Information" TUTORIAL Title: "Privacy preserving data sharing and mining" Presenters: Assoc Prof Jiuyong Li, Dr Peter Christen, and Dr Artak Amirbekyan Topics: 1) Introduction to privacy and data sharing 2) Data anonymisation 3) Privacy preserving data linkage 4) Secure Multiparty-Computation based data mining We look forward to see you at AusDM'08 in Glenelg. Best regards, AusDM'08 chairs. Contact: ausdm08@togaware.com Useful Links and Publications Useful links for those readers interested in keeping up with developments in data mining: Books and Journals on Data Mining http://www.the-data-mine.com/bin/view/Misc/DataMiningBooksAndPapers KDNuggets This site is operated by Gregory Piatetsky-Shapiro. Gregory is an active member of SIGKDD in the USA. www.kdnuggets.com Togaware This site is run by Dr Graham Williams from IAPA and is a popular data-mining site. www.togaware.com Credit Risk Analytics Occasional Newsletter This is a publication produced by Dr Ross Gayler a senior research and development consultant with Veda Advantage and an active member of IAPA. Next Edition of the IAPA Newsletter - Issue 34 This will cover post graduate education in analytics Newsletter editor is Warwick Graco Vice Chair for Communications for IAPA Please notify IAPA of any forthcoming and important industry events on iapa.org@gmail.com or Anthony.Nolan@ato.gov.au If you would like to contribute to IAPA Newsletter, contact us on iapa.org@gmail.com IAPA NEWSLETTER is a fortnightly publication of the Institute of Analytics Professionals Australia and is provided to notify members of key issues and developments. This includes messages from the IAPA Board, notifications of events, vacant positions, short clips on personalities and new developments with tools and methods. Please feel free to circulate these newsletters to other interested parties. Copyright © 2007 IAPA. |
|
| |
|
16th April 2009, 09:16 AM
| #3 |
| Administrator | IAPA Newsletter - Issue 35 26 March 2009 IAPA Newsletter - Issue 35 In today's Newsletter:
Announcements IAPA announces the appointment of new Chapter Heads for NSW and Victoria. The Chair of IAPA, Doug Campbell, stated "We are very pleased to announce the appointment of two high profile industry leaders for our largest Chapters. Both Carlie and Drew are well known for their success in the field of analytics and will bring strong ideas and networks to IAPA. By strengthening our Industry involvement, IAPA will continue to enhance the benefits that our members seek and increase our relevance to our partners and sponsors. With leading organisations NAB and Vanguard joining the Australian Tax Office in providing local IAPA leadership in Sydney, Melbourne and Canberra respectively, our membership and profile will continue to grow". New Head of the Victorian Chapter Andrew (Drew) Harvey has been appointed to the Head of the Victorian Chapter for IAPA (taking over from Doug Campbell who moved on to the Chairman's position). Drew's first event in Melbourne in February saw a record attendance hear from Tony Woods, an acclaimed speaker on Analytics internationally. Drew has long been active in the field of Analytics and will bring strong `not-for-profit' experience to assist with the running of IAPA. A profile of Drew follows: Drew is currently Manager - Client Insight for Vanguard Investments Australia. Leading a new initiative for Vanguard, Drew is the business owner for the CRM and Business Intelligence strategy, and is responsible for establishing an analytics capability across the Personal Investor and Adviser markets to facilitate significant sales growth in Retail and Wholesale Index Funds Management. Drew's previous roles have included: Marketing Analytics Manager for Coles Group: Drew established a B2C sales model, using data mining and analytics to provide FMCG suppliers of Coles Group with an avenue to communicate directly to FlyBuys customers exhibiting specific consumer behaviours. This delivered increased sales across many categories such as Health & Beauty, Pet foods, Dairy, Personal Care, and Grocery Fundraising Manager for The Cancer Council Victoria: Drew directed a number of fundraising initiatives, and was responsible for generating in excess of 50% of the Council's annual income. He generated a dramatic increase in direct mail income through the use of segmentation of donors and testing of direct mail material. Marketing Director (Fundraising) for The University of Cambridge: Drew was responsible for all mass fundraising activities for the University of Cambridge, including a global direct marketing initiative, and establishing outbound call centres in Cambridge (UK) and Toronto (Canada). During his time at Cambridge, annual income from donors increased 208%. Drew has a first-class Honours degree in Maths and Stats from Leicester University UK, and has previously held actuarial, credit control and trade settlement roles for Financial Services companies in London. New Head of the NSW Chapter Carlie Smee has been appointed to the Head of the NSW Chapter, taking over from Eugene Dubossarsky who has stepped down to focus on Mentoring. In 2006, Carlie was appointed as the Head of Customer Management and Analytics for nab's start up direct banking business UBank. To date this role has involved establishing the customer management and analytics team and capability from scratch. Going forward focus will be on delivering best in class analytics capability, marketing optimisation and driving profit for UBank by delivering ROI from customers via customer management initiatives such as cross sell, customer retention and touch point management. Carlie Smee has a proven track record leading research and analytical teams across multiple industries, primarily financial services. Prior to her position at nab, Carlie ran the Marketing Services team for Virgin which included the functions of analytics, market research, marketing delivery and partner management. Carlie has arranged for the Commonwealth Bank (CBA) to speak at the next NSW IAPA Forum - confirmed for Monday 27th April 2009. CBA are recognised as leaders in the credit and risk management field. Three senior members of CBA's Risk team will jointly present a case study on techniques and strategies CBA apply to effectively manage this business unit. Logistics: From 5.30pm - Registration 6pm - Welcome by the new Head of NSW 6.10pm - Presentation and interactive discussion 7pm - 8pm- Networking over drinks and canapés Venue: Deloitte, L9, 225 George Street, Sydney RSVP : rsvp@iapa.org.au Speaker details: Jonathan Passlow joined CBA in 2004, running reporting and originations for personal loans and now credit cards. During this time, Jonathan has optimised origination strategies for both credit cards and personal loans, implemented the monitoring of credit risk factors, as well as playing a key role in the development of contribution modelling. Andrew Fairthorne was recruited as a Senior Analyst in CBA's credit card account management unit. Having joined CBA in 2007 Andrew has played an important role in driving the optimisation of the credit card authorisation system, which has contributed to improved credit risk assessment at a transactional level, and also enhanced customer service at point of sale. Prior to CBA, Andrew was a quantitative analyst with the Commonwealth Department of Employment, Education and Workplace Relations. Paul Fishburn joined CBA in 2006 to manage cards collections strategy for CBA. During this time, Paul has represented credit risk in the collections transformation project and developed collections strategies that piloted new contact channels. Prior to CBA Paul was based in the UK working for Marks & Spencer Financial Services (UK) in credit card account management. Editorial There was an interesting article by Joe McKendrick in a recent Teradata IQ Newsletter on automating operational decisions. The article argues that big decisions involving policies, strategy and other key issues will remain the responsibility of senior executives while many high-volume operational and tactical decisions can be automated. These applications with repeatable outcomes consist of business rules and other routines and are applied to incoming data. The applications are regularly updated based on the results of previous successes and failures. They can protect organisations from fraud and other intrusions, save countless employee hours that can be wasted on routine matters and reduce time to respond to market opportunities. One important class of tools that will make this automated decision making possible is referred to as knowledge acquisition and decision engineering or KADE. Expert knowledge is required to drive effective decision making while decision engineering is required to ensure the decisions taken are optimal (ie free of unwanted biases and other sources of error and achieve the best possible outcomes). There are many areas where automated decision making can be applied. Some obvious examples include: traffic lights to improve the flow of vehicles, reduce driver frustration and fatigue and lower pollution levels black box technologies used in cars, homes, offices and places of entertainment to minimise energy consumption loan applications that are relatively straight forward leaving more difficult ones to staff to process It is suggested that automated decision making will be a growth area for our profession. http://www.teradata.com/tdmo/v08n04/...andInHand.aspx Masters Program in Analytics The Board at IAPA has continued discussions with a number of academic institutions including ANU and James Cook University. One insight that came from these discussions and was highlighted in Newsletter No 34 was that Analytics is faced with a demand rather than supply problem. Academic institutions are prepared to provide programs if the demand is there from students. This led to the observation that Analytics is poorly marketed at present. Managers do not know much about it and the same applies to high school students who are making career choices. Both need to be targeted and this is especially so in the current economic climate. Intelligence gained from those who work in data warehousing and in analytics software indicates that the demand for those who can do analysis increases in recessions as firms look for windows of opportunity to lower costs and to increase profits. The next steps are to raise awareness of analytics amongst managers and students so that demand for these skilled specialists grows. IAPA is working on initiatives aimed at addressing these gaps. One is a booklet which will give managers an introduction to analytics. This booklet is currently being written by members of the Institute. Another initiative is to write a letter to university vice chancellors to bring to their attention this emerging discipline. A third one is to target students in high schools to increase their understanding that there are not only careers in fields such as law, science and engineering; but also in the knowledge and analytical disciplines. Meta Learning One of the major challenges facing model developers is selecting the best algorithm(s) based on the problems being solved, the features available that describe the instances in the population of interest and the metrics used to assess the performance of the models. It has been long been recognised that there is no universally best algorithm for all problem domains. There is also no gold standard for selecting algorithms. What is required is firstly to understand the nuances of the problem to be solved, secondly the features that can be used as inputs to the model and thirdly the limitations of the data that is available. One researcher and academic who has reviewed developments with algorithm selection is Professor Smith-Miles Head of the School of Engineering and Information Technology at Deakin University, Australia. She has approached this problem from the perspective of meta-learning. The goal of this learning is to model the relationships between the performance of various learning algorithms and the characteristics of problems being learned. The focus is on learning about learning such as under what conditions can one expect a certain algorithm to perform well? This approach has been very well developed in the machine learning community since the early 1990s. While it is hazardous to try to predict the future, it is expected that giant strides will not be made in algorithm selection until meta-knowledge is integrated with the algorithms. Model developers currently use `dumb' algorithms where their application is determined by the judgments of the model developers. Meta-knowledge captures the expert thinking and learning on the data used and the algorithms applied to classify and predict. This knowledge guides the selection of algorithms and adjustments are made based on feedback obtained from algorithm performance. This knowledge needs to be continually updated as issues are examined and algorithms are tested. This subject will be revisited in the future as more is learned about meta-learning and algorithm selection. Three papers published by Professor Smith-Miles on meta-learning include: Kate Smith-Miles (2008) Cross-Disciplinary Perspectives on Meta-Learning for Algorithm Selection. ACM Computing Surveys (CSUR). Volume 41,Issue 1, (December 2008) Kate Smith-Miles (2008). Towards insightful algorithm selection for optimisation using meta-learning concepts. In: International Joint Conference on Neural Networks, pp. 4117--4123 Kate Smith-Miles (2007). Meta-Learning: From Classification to Forecasting, to Optimization, and Beyond. icis,pp.2, 6th IEEE/ACIS International Conference on Computer and Information Science (ICIS 2007) Interview with Chief Economist at Yahoo This and the next item were supplied by Dr Rohan Baxter Hal Varian is Chief Economist at Google and in an interview supplied the following response to the question below Q: Your job sounds extremely interesting. What jobs would you recommend to a young person with an interest, and maybe a bachelors degree, in economics? A: If you are looking for a career where your services will be in high demand, you should find something where you provide a scarce, complementary service to something that is getting ubiquitous and cheap. So what's getting ubiquitous and cheap? Data. And what is complementary to data? Analysis. So my recommendation is to take lots of courses about how to manipulate and analyze data: databases, machine learning, econometrics, statistics, visualization, and so on. The interview can be found at: http://freakonomics.blogs.nytimes.co...ons/#more-2345 A Longstanding Principle in Statistics The following blog is from Andrew Gelman Hal Pashler writes: I [Pashler] thought you guys would enjoy this charming little 1950 paper by Edward Cureton entitled "Validity, Reliability, and Baloney" (Dirk Vorberg, a German math psych guy, sent it). Long before machine learning, it seems that psychometrics people were confronting this issue--and the concrete form it took was "What should we make of validation measures computed with the same data that were used to select out particular items for inclusion in the test?". Just swap voxels for items, and it's the same problem [as in the Vul, Harris, Winkelman, and Pashler paper on suspiciously high correlations in bran imaging studies]. This reminds me of a longstanding principle in statistics, which is that, whatever you do, somebody in psychometrics already did it long before. I've noticed this a few times. Once, about ten years ago, I was at a conference where computer scientists were talking about some pretty elaborate statistical models, and I realized these were the same as some things I'd seen Iven Van Mechelen and his colleagues working on in the Psychology Department at Leuven. Then, more recently, I wrote this article with David Park on splitting a predictor into three parts, and it turned out that similar work had been done in 1928! by psychometric researcher T. L. Kelley (and, oddly enough, E. Cureton in 1957). http://www.stat.columbia.edu/~cook/m...gstanding.html http://www.stat.columbia.edu/~gelman...on_Baloney.pdf http://www.stat.columbia.edu/~gelman...ed/thirds5.pdf New Books Knowledge Discovery for Counterterrorism and Law Enforcement By David Skillicorn Hardcover: 352 pages Publisher: CRC; 1 edition (November 13, 2008) Language: English ISBN-10: 1420073990 ISBN-13: 978-1420073997 This book provides a refreshing look at the use of knowledge discovery in intelligence, security and law enforcement. It focuses on four main forms of knowledge discovery including prediction, clustering, relationship discovery, and textual analysis. For each of these application areas, the author discusses opportunities for concealment and cover that are available to criminals and subversives and reveals some of the tactics that can aid in detecting them. He reviews what is known about the different technologies for each area and evaluates their effectiveness. The book also reviews technologies currently under development and describes how they will integrate with existing approaches to knowledge discovery. While all discovery and detection systems are susceptible to manipulation, those armed with the knowledge of these tactics are better able to create systems to avoid these vulnerabilities. This book delineates an effective process for integrating knowledge-discovery tools, provides a unique understanding of the limits of the technology, and contains a clear presentation of the upsides and pitfalls of data collection. It is a useful addition in the library of those who have to identify the sophisticated tactics employed by criminals and other unsavoury individuals and organizations. http://www.taylorandfrancis.com/shop...h%3Dskillicorn www.amazon.com/Knowledge-Discovery-Counterterrorism-Enforcement-Chapman/dp/1420073990/ref=sr_1_5?ie=UTF8&s=books&qid=1226004467&sr=8-5" The Genie in the Machine: How Computer-Automated Inventing is Revolutionizing Law and Business by Robert Plotkin Published by Stanford University Press Spring 2009 ISBN 978-0804756990 We have long considered inventing to be a uniquely human activity. But just as the assembly line automated the process of manufacturing, today's computers are automating the process of inventing. Software can automatically generate designs for everything from toothbrushes to antennas to automobile frames more quickly and inexpensively than ever before, thereby ushering in a new era of artificial invention. Inventors will use artificial invention technology to boost their inventive abilities to previously undreamed-of heights, enabling small teams of inventors to compete with mega-corporations who insist on inventing the old-fashioned way. Even consumers will be able to use artificial invention technology to become inventors themselves. We stand poised to see the emergence of the "digital renaissance artisan" -- a person who will have the ability to not only design new inventions at the touch of a button, but also to manufacture them automatically from the comfort of home. http://www.automatinginvention.com/t...e-machine.html Notices Report on the 7th Australasian Data Mining Conference (AusDM 2008), Glenelg, South Australia, 27 and 28 November 2008. The following summary is supplied by Dr Peter Christen from ANU. AusDM'08 in Glenelg was a great success thanks to the great team work of the organisation committee John Roddick (Flinders), Jiuyong Li (UniSA), Peter Christen (ANU), Paul Kennedy (UTS), Simeon Simoff (Western Sydney) and Graham Williams (ATO). We had 45 attendants from both academia and industry, including a number of overseas academics from New Zealand, Korea, Thailand, Singapore, China, the USA, Canada, France, and Spain, who presented their research. Industrial attendees were from the ATO, Inland Revenue Department (New Zealand), Medicare Australia, GMHBA, AUSTRAC, LG CNS Co. Ltd. (Korea), the Australian Antarctic Division, Virid Pty Ltd, Delta Electricity, DSTO, and DSO National Laboratories (Singapore). Three exciting keynote speaker gave talks about various aspects of data mining and analytics: - Professor David Powers presented a colourful multi-media talk about his computational research on how children learn, and he demonstrated several interesting applications of his research. - Professor Geoff Webb talked about multi-strategy ensemble learning, and he made the organisation committee dance in front of the audience to illustrate the advantages of ensemble learning. - Dr Richard Price discussed the challenges intelligence agencies have when trying to make sense of the large volumes of intelligence information, including multimedia and free-text data, that needs to be analysed in near real-time. There were 21 regular presentations, selected from 55 submitted papers. New at AusDM'08 was a Doctoral Consortium where PhD students in their first year presented their research ideas, and a tutorial on Privacy-Preserving Data Sharing and Mining. Both proved to be a great success and it is planned to again include such special sessions at this year's AusDM'09. All papers presented at AusDM'08 are now available for download on the Conferences in Research and practice in Information Technology (CRPIT) at: http://crpit.com/Vol87.html AusDM will continue in 2009, and as in previous years again be held together with the Australasian Joint Conference on Artificial Intelligence (AI'09), in Melbourne in early December 2009. For more details on AI'09 please see: http://www.infotech.monash.edu.au/ai09 The papers for AusDM 08 can be downloaded from: http://crpit.com/Vol87.html Open Source in Data Mining workshop (OSDM09) The preliminary program for upcoming Open Source in Data Mining workshop (OSDM09), to be held at the Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD'09) in Bangkok in April 2009. OSDM'09 Web site: http://osdm09.togaware.com/ REGISTRATION Early registration discount is available until the end of February using the PAKDD online registration system at: http://itpe.siit.tu.ac.th/pakdd2009/...w/registration International Workshop on Social Computing, Behavioral Modeling, and Prediction Phoenix, Arizona March 31-April 1, 2009 Call for Participation - Social Computing, Behavioral Modeling, and Prediction (SBP09) Early Registration by March 1, 2009 Proceedings by Springer available at http://www.amazon.com/Social-Computi.../dp/1441900551 Invited Speakers: http://www.public.asu.edu/~huanliu/s...edspeaker.html Program: http://www.public.asu.edu/~huanliu/sbp09/program.html Registration and Hotel Information: http://www.public.asu.edu/~huanliu/s...istration.html Supported by AFOSR, ARL, ONR, sponsored by NSF, and NIH (pending) Notifications of Other Conferences Some of these are found at the links for KDNuggets and for Credit Risk Analytics Occasional Newsletter below Useful links and Publications Useful links for those readers interested in keeping up with developments in data mining: Books and Journals on Data Mining http://www.the-data-mine.com/bin/vie...BooksAndPapers KDNuggets This site is operated by Gregory Piatetsky-Shapiro. Gregory is an active member of SIGKDD in the USA. www.kdnuggets.com Togaware This site is run by Dr Graham Williams from IAPA and is a popular data-mining site. www.togaware.com Credit Risk Analytics Occasional Newsletter This is a publication produced by Dr Ross Gayler a senior research and development consultant with Veda Advantage and an active member of IAPA. Ross.Gayler@VedaAdvantage.com. Next Edition of the IAPA Newsletter - Issue 36 This will cover anomaly detection Newsletter editor is Warwick Graco Vice Chair for Communications for IAPA Please notify IAPA of any forthcoming and important industry events on iapa.org@gmail.com or Anthony.Nolan@ato.gov.au If you would like to contribute to IAPA Newsletter, contact us on iapa.org@gmail.com IAPA NEWSLETTER is a fortnightly publication of the Institute of Analytics Professionals Australia and is provided to notify members of key issues and developments. This includes messages from the IAPA Board, notifications of events, vacant positions, short clips on personalities and new developments with tools and methods. Please feel free to circulate these newsletters to other interested parties. |
|
| |
|
| Bookmarks |
| Thread Tools | Search this Thread |
| Display Modes | |
| |
Similar Threads | ||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Sybase Analytic Appliance | Doug Heywood | IBM and Cognos Forum | 0 | 11th November 2008 09:08 AM |
| IAPA Forecasting Special Interest Group Meeting (FSIG) on September 3 | binboy | Analytics | 0 | 20th August 2008 10:58 AM |
| IAPA Forecasting Special Interest Group | admin | Dashboarding and Scorecard Tips and Techniques | 0 | 6th August 2008 05:12 PM |
| Institute of Analytics Professionals of Australia | Steve Bennett | Local Happenings | 0 | 15th October 2007 11:07 PM |
| | |
| | |
Linear Mode
