1 / 41

Literature Network

Honorary Doctoral Degree for Prof. Dr. Lennart Ljung and Workshop on System Identification and Data Modeling Wednesday October 12 - 13, 2004 Katholieke Universiteit Leuven. Literature Network.

Download Presentation

Literature Network

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Honorary Doctoral Degree forProf. Dr. Lennart Ljung and Workshop on System Identification and Data ModelingWednesday October 12 - 13, 2004Katholieke Universiteit Leuven

  2. Literature Network

  3. Hierarchical text-based clustering of the 138 seed papersDendrogram: all papers clustered into 4 groups The algorithm was able to separate most of the 13 papers belonging to another ‘Ljung L’. Other important words in the suspicious cluster are uterine, electron microscopy, epithelium, … The 13 corresponding seed papers were removed from the dataset Answer to question 5

  4. Hierarchical text-based clustering of the 125 seed papersDendrogram: all papers clustered into 2 groups

  5. Hierarchical text-based clustering of the 125 seed papersDendrogram: all papers clustered into 4 groups Answer to question 6:

  6. Hierarchical text-based clustering of the 125 seed papersDendrogram: all papers clustered into 6 groups

  7. ‘identifiability’ Cluster Term Network (4 papers)

  8. ‘architectures’ Cluster Term Network (22 papers)

  9. ‘models’ Cluster Term Network (65 papers)

  10. ‘transfer functions’ Cluster Term Network (4 papers)

  11. ‘convergence’ Cluster Term Network (19 papers)

  12. ‘algorithms’ Cluster Term Network (11 papers)

  13. Multidimensional Scaling, 125 papers: 2DDifferent color-symbol combination for each of 6 clusters

  14. Multidimensional Scaling, 125 papersEigenvalues of Y*Y’ (with Y “configuration matrix”)

  15. Multidimensional Scaling, 125 papers: 2DDifferent color-symbol combination for each year of publication

  16. Binning papers using sliding window of 5 years 1991-1995: 22 1992-1996: 26 1993-1997: 24 1994-1998: 23 1995-1999: 23 1996-2000: 23 1997-2001: 22 1998-2002: 26 1999-2003: 28 2000-2004: 25 ALL: 125 Number of papers in each 5 year bin: 1972-1976: 15 1973-1977: 19 1974-1978: 27 1976-1980: 26 1977-1981: 20 1978-1982: 19 1979-1983: 12 1980-1984: 12 1981-1985: 19 1982-1986: 21 1983-1987: 18 1984-1988: 18 1985-1989: 16 1986-1990: 9 1987-1991: 8 1988-1992: 13 1989-1993: 14 1990-1994: 17 1975-1979: 29 Most active period (Q3.1)

  17. Hierarchical text-based clusteringDendrogram: most active period (1975-1979)one paper per leaf

  18. Hierarchical text-based clusteringDendrogram: most active period (1975-1979)cut-off at 5 groups Answer to question 3.2: “adaptive” is least relevant

  19. Multidimensional Scaling, 125 papers: 2DBest term or phrase for each paper published in the indicated 5 year window

  20. Multidimensional Scaling, 125 papers: 2DBest term or phrase for each paper published in the indicated 5 year window

  21. Hierarchical text-based clustering of the citing papersDendrogram: all papers clustered into 20 groups

  22. Term Network for all citing papers

  23. Hierarchical text-based clustering of available cited papersDendrogram: all papers clustered into 20 groups

  24. Term Network for all available cited papers

  25. Hierarchical text-based clustering of Lennart Ljung papersw.r.t. complete 1st degree literature networkDendrogram: all papers clustered into 20 groups

  26. Term Network for all papers of Lennart Ljungw.r.t. complete 1st degree literature network

  27. Co-authorship network, authors with “Ljung Number 1” Answer to question 1.1: - Total number of different co-authors: 77 • Total number of co-authors on all papers: 175 • “collaboration degree” = 175/125 = 1.4

  28. Citation statistics • - Up till July 2004 Lennart Ljung has been cited 3531 times. • Number of citations per paper in descending order (July 2004) : Answer to question 2: 500 citations • Mean citation rate is 28.2 and the median is 7

  29. 10 most cited papers according to WoS[times cited]: [authors] - [title] 500: LJUNG L - ANALYSIS OF RECURSIVE STOCHASTIC ALGORITHMS 234: ASTROM KJ, BORISSON U, LJUNG L, WITTENMARK B - THEORY AND APPLICATIONS OF SELF-TUNING REGULATORS 196: LJUNG L - POSITIVE REAL TRANSFER-FUNCTIONS AND CONVERGENCE OF SOME RECURSIVE SCHEMES 178: LJUNG L - ASYMPTOTIC-BEHAVIOR OF THE EXTENDED KALMAN FILTER AS A PARAMETER ESTIMATOR FOR LINEAR-SYSTEMS 173: Sjoberg J, Zhang QH, Ljung L, Benveniste A, Delyon B, Glorennec PY, Hjalmarsson H, Juditsky A - Nonlinear black-box modeling in system identification: A unified overview 128: GUSTAVSSON I, LJUNG L, SODERSTROM T - IDENTIFICATION OF PROCESSES IN CLOSED-LOOP - IDENTIFIABILITY AND ACCURACY ASPECTS 125: SODERSTROM T, LJUNG L, GUSTAVSSON I - THEORETICAL-ANALYSIS OF RECURSIVE IDENTIFICATION METHODS 122: LJUNG L - CONVERGENCE ANALYSIS OF PARAMETRIC IDENTIFICATION METHODS 115: FALCONER DD, LJUNG L - APPLICATION OF FAST KALMAN ESTIMATION TO ADAPTIVE EQUALIZATION 109: LJUNG L, MORF M, FALCONER D - FAST CALCULATION OF GAIN MATRICES FOR RECURSIVE ESTIMATION SCHEMES

  30. Source titles Answer to question 7

  31. Citation statistics

  32. Citation statistics

  33. Citation statisticsTotal number of citations given to any paper of Lennart Ljung in each year

  34. References in papers of Lennart Ljung • Lennart Ljung has cited 2732 papers in total in his 125 papers known to WoS • Number of references in descending order: • The mean number of references is 21.9 and the median is 17

  35. Ljung’s favorite authors [times cited]: [author1, author2,...] 188: LJUNG L 63: KAILATH T 46: SODERSTROM T 29: GUSTAVSSON I 24: MORF M 22: GUO L 20: ASTROM KJ 18: YUAN ZD, ANDERSON BDO 17: MACCHI O, GOODWIN GC 16: HJALMARSSON H, BENVENISTE A, WAHLBERG B, FRIEDLANDER B 15: EWEDA E 14: GEVERS M, BITMEAD RR 13: GUNNARSSON S, VANDENHOF PMJ, PRIOURET P 11: GARDNER WA, LAINIOTI.DG, NINNESS B 10: KUSHNER HJ, SCHRAMA RJP, HANNAN EJ 9: JUDITSKY A, GLOVER K, BILLINGS SA, STOICA P, DELYON B, SJOBERG J 8: DEMOOR B, KOSUT RL, WIDROW B, ZHANG Q, LANDAU ID, AKAIKE H 7: CAINES PE, VERHAEGEN M, NARENDRA KS, POOLLA K, GUSTAFSSON F 6: WILLEMS JC, HUANG H, GLORENNEC PY, KHARGONEKAR PP, SIDHU GS, BREIMAN L, MCCOOL JM, SARIDIS GN, WITTENMARK B, JOHNSON CR, BERK KN, WITTENMA.B, RISSANEN J, FORSSELL U 5: KRAUSE J, MORARI M, STONE CJ, NAGPAL K, JUANG JN, GARULLI A, BEMPORAD A, FARDEN DC, BORISSON U, ISERMANN R, MAREELS MY, NIEDZWIECKI M, VANOVERSCHEE P, KABAILA PV, LARIMORE MG, KHARGONEKAR P, MILANESE M, TIKKU A, LINDQUIST A, DEWILDE P, VICINO A ...

  36. Ljung’s favorite authors, excluding co-authors [times cited]: [author1, author2,...] 17: MACCHI O, GOODWIN GC: Answer to question 4 15: EWEDA E 14: BITMEAD RR 13: VANDENHOF PMJ 11: GARDNER WA, LAINIOTI.DG, NINNESS B 10: KUSHNER HJ, SCHRAMA RJP, HANNAN EJ 9: BILLINGS SA, STOICA P 8: DEMOOR B, KOSUT RL, WIDROW B, LANDAU ID, AKAIKE H 7: CAINES PE, NARENDRA KS, POOLLA K 6: WILLEMS JC, HUANG H, KHARGONEKAR PP, SIDHU GS, BREIMAN L, MCCOOL JM, SARIDIS GN, JOHNSON CR, BERK KN, WITTENMA.B, RISSANEN J 5: KRAUSE J, MORARI M, STONE CJ, NAGPAL K, JUANG JN, FARDEN DC, ISERMANN R, MAREELS IMY, NIEDZWIECKI M, VANOVERSCHEE P, KABAILA PV, LARIMORE MG KHARGONEKAR P, MILANESE M, TIKKU A, LINDQUIST A, DEWILDE P, VICINO A …

  37. Ljung’s favorite papers, excluding own[times cited]: [authors] - [title] 10: KAILATH T - SOME NEW ALGORITHMS FOR RECURSIVE ESTIMATION IN CONSTANT LINEAR-SYSTEMS 7: GARDNER WA - NONSTATIONARY LEARNING CHARACTERISTICS OF THE LMS ALGORITHM 7: EWEDA E, MACCHI O - TRACKING ERROR-BOUNDS OF ADAPTIVE NONSTATIONARY FILTERING 6: MACCHI O, EWEDA E - 2ND-ORDER CONVERGENCE ANALYSIS OF STOCHASTIC ADAPTIVE LINEAR-FILTERING 6: BERK KN - CONSISTENT AUTOREGRESSIVE SPECTRAL ESTIMATES 6: LAINIOTI.DG - PARTITIONED ESTIMATION ALGORITHMS .2. LINEAR ESTIMATION 5: WIDROW B, MCCOOL JM, LARIMORE MG, JOHNSON CR - STATIONARY AND NONSTATIONARY LEARNING CHARACTERISTICS OF LMS ADAPTIVE FILTER 5: SARIDIS GN - COMPARISON OF 6 ON-LINE IDENTIFICATION ALGORITHMS 5: ASTROM KJ, WITTENMA.B - SELF TUNING REGULATORS 5: VANDENHOF PMJ, SCHRAMA RJP - AN INDIRECT METHOD FOR TRANSFER-FUNCTION ESTIMATION FROM CLOSED-LOOP DATA 5: POOLLA K, KHARGONEKAR P, TIKKU A, KRAUSE J, NAGPAL K - A TIME-DOMAIN APPROACH TO MODEL VALIDATION 4: VERHAEGEN M - IDENTIFICATION OF THE DETERMINISTIC PART OF MIMO STATE-SPACE MODELS GIVEN IN INNOVATIONS FORM FROM INPUT-OUTPUT DATA 4: KUSHNER HJ, HUANG H - ASYMPTOTIC PROPERTIES OF STOCHASTIC APPROXIMATIONS WITH CONSTANT-COEFFICIENTS 4: ASTROM KJ - MAXIMUM-LIKELIHOOD AND PREDICTION ERROR METHODS …

  38. Global trails Answer to question 1.2: 4

  39. Other solutions Answer to question 8: Lennart Ljung was born on SUNDAY September 13th, 1946 Answer to question 9:

  40. Contact: frizo.janssens@esat.kuleuven.ac.be Thank you: Dries Van Dromme Patrick Glenisson Bart De Moor

  41. References(Biolayout, 2001) A.J. Enright, C.A. Ouzounis, BioLayout JAVA, Bioinformatics, 2001, 17/ 853-854, http://maine.ebi.ac.uk:8000/services/biolayout/(van Dongen, 2000) van Dongen S., “A cluster algorithm for graphs”, Technical Report INS-R0010, National Research Institute for Mathematics and Computer Science in the Netherlands, Amsterdam, May 2000, http://www.cwi.nl/ftp/CWIreports/INS/INS-R0010.ps.Z

More Related