000 04176nam a22004935i 4500
001 978-3-319-01321-3
003 DE-He213
005 20140220082840.0
007 cr nn 008mamaa
008 131203s2013 gw | s |||| 0|eng d
020 _a9783319013213
_9978-3-319-01321-3
024 7 _a10.1007/978-3-319-01321-3
_2doi
050 4 _aQA71-90
072 7 _aPDE
_2bicssc
072 7 _aCOM014000
_2bisacsh
072 7 _aMAT003000
_2bisacsh
082 0 4 _a004
_223
100 1 _aPaprotny, Alexander.
_eauthor.
245 1 0 _aRealtime Data Mining
_h[electronic resource] :
_bSelf-Learning Techniques for Recommendation Engines /
_cby Alexander Paprotny, Michael Thess.
264 1 _aCham :
_bSpringer International Publishing :
_bImprint: Birkhäuser,
_c2013.
300 _aXXIII, 313 p. 100 illus., 12 illus. in color.
_bonline resource.
336 _atext
_btxt
_2rdacontent
337 _acomputer
_bc
_2rdamedia
338 _aonline resource
_bcr
_2rdacarrier
347 _atext file
_bPDF
_2rda
490 1 _aApplied and Numerical Harmonic Analysis,
_x2296-5009
505 0 _a1 Brave New Realtime World – Introduction -- 2 Strange Recommendations? – On The Weaknesses Of Current Recommendation Engines -- 3 Changing Not Just Analyzing – Control Theory And Reinforcement Learning -- 4 Recommendations As A Game – Reinforcement Learning For Recommendation Engines -- 5 How Engines Learn To Generate Recommendations – Adaptive Learning Algorithms -- 6 Up The Down Staircase – Hierarchical Reinforcement Learning -- 7 Breaking Dimensions – Adaptive Scoring With Sparse Grids -- 8 Decomposition In Transition - Adaptive Matrix Factorization -- 9 Decomposition In Transition Ii - Adaptive Tensor Factorization -- 10 The Big Picture – Towards A Synthesis Of Rl And Adaptive Tensor Factorization -- 11 What Cannot Be Measured Cannot Be Controlled - Gauging Success With A/B Tests -- 12 Building A Recommendation Engine – The Xelopes Library -- 13 Last Words – Conclusion -- References -- Summary Of Notation.
520 _aDescribing novel mathematical concepts for recommendation engines, Realtime Data Mining: Self-Learning Techniques for Recommendation Engines features a sound mathematical framework unifying approaches based on control and learning theories, tensor factorization, and hierarchical methods. Furthermore, it presents promising results of numerous experiments on real-world data.  The area of realtime data mining is currently developing at an exceptionally dynamic pace, and realtime data mining systems are the counterpart of today's “classic” data mining systems. Whereas the latter learn from historical data and then use it to deduce necessary actions, realtime analytics systems learn and act continuously and autonomously. In the vanguard of these new analytics systems are recommendation engines. They are principally found on the Internet, where all information is available in realtime and an immediate feedback is guaranteed.   This monograph appeals to computer scientists and specialists in machine learning, especially from the area of recommender systems, because it conveys a new way of realtime thinking by considering recommendation tasks as control-theoretic problems. Realtime Data Mining: Self-Learning Techniques for Recommendation Engines will also interest application-oriented mathematicians because it consistently combines some of the most promising mathematical areas, namely control theory, multilevel approximation, and tensor factorization.
650 0 _aMathematics.
650 0 _aComputer science.
650 0 _aComputer software.
650 1 4 _aMathematics.
650 2 4 _aComputational Science and Engineering.
650 2 4 _aMathematical Applications in Computer Science.
650 2 4 _aMathematical Software.
700 1 _aThess, Michael.
_eauthor.
710 2 _aSpringerLink (Online service)
773 0 _tSpringer eBooks
776 0 8 _iPrinted edition:
_z9783319013206
830 0 _aApplied and Numerical Harmonic Analysis,
_x2296-5009
856 4 0 _uhttp://dx.doi.org/10.1007/978-3-319-01321-3
912 _aZDB-2-SMA
999 _c96534
_d96534