                           ˿µϣ20130113գ

13.01.09-12ûи
13.01.13, йѧԺԺʿӻ׽˿ѧ
13.01.13, ӢѧѧDavid Willetts·ӵĺ
13.01.13, ̸ӷ̸۳ʱ
13.01.13, Ѻ΢̸̸жܷҩ
13.01.13, ΢רϷ԰ڣİȫ
13.01.13, ʡ¶ȣߵ˵硷
13.01.13, йУҪķʿƸĸ
13.01.13, ȺԡϺпѧѧоʱ١
13.01.13, ϰ˼ϺпѧѧоԼЪ˹
13.01.13, ʤƺӡ¡ʺذذӰ졷һƪϮ¡
13.01.13, ɵڡƴšиڱƼѧƸУʧƫġ
13.01.13, ⽨ΡԡϾѧѧƸ־־ΪѧԵǸĳ塷
13.01.13, Ŷ»ƵԸס
13.01.13, 360͵׬Ӷ𣿲Ա꿪ʼ

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

йѧԺԺʿӻ׽˿ѧ

׽˿ѧйѧԺԺʿооԱӣ
Աڷαѧа̺αҽռƹ
ԴȷĹס

佱ʽ20131132ڱʱУԺʿڷ
񽱸Ժ˹й˵翪ı档

˿ѧڽڰйѧͻ׵
ʿÿѡһΣÿһ·ݰ䷢һˣһʿɡý
˿ίѡλʿĿѧڿ湫˾MDPI˾
mdpi.com

߼飺ӣ1927Ϻ1951ҵ廪ѧϵ
19591960Ųоμо19611965
ҹԭӵⵯо19651966μ˲ģ͵оǲ
ģ͵ĸҪ֮һ19741979оӳۡ1980֮
Ӻо1980굱ѡйѧԺѧίԱԺʿ
ĿǰΪйѧԺооԱͰˡʮ¶
칦ܡҩˮ͡αѧ1999а̡
ԾڷαҽƹԴ

ӣTso-hsiu HoѧӢ嵥

SEARCH FOR NEW MASSIVE PARTICLES IN COSMIC-RAYS 
CHEN HS, DAI CJ, DING LK, GUO YN, HUO AX, JING CL, KUANG HH, LIU HT, MA 
JM, SHEN CQ, SHENG HY, YAO ZG, YU ZQ, ZHU QQ,  Ching CQ, HO TH, GAO CS
PHYSICS REPORTS-REVIEW SECTION OF PHYSICS LETTERS , v. 282(1) pp. 2-34 APR 1997 

A POSSIBLE EXPLANATION OF THE NEGATIVE VALUES OF M(NU-E)(2) OBTAINED FROM THE BETA-SPECTRUM SHAPE ANALYSES 
CHING CR, HO TH 
INTERNATIONAL JOURNAL OF MODERN PHYSICS A , v. 10(19) pp. 2841-2849 JUL 30, 1995 

OPERATOR EXPANSION METHOD AND NUCLEAR BETA-BETA-DECAY 
HIRSCH M, WU XR, KLAPDORKLEINGROTHAUS HV, CHING CR, HO TH 
PHYSICS REPORTS-REVIEW SECTION OF PHYSICS LETTERS , v. 242(4-6) pp. 403-422 JUL 1994 

NEW THEORETICAL RESULTS OF 2-NU-BETA-BETA DECAY WITH THE OPERATOR EXPANSION METHOD 
WU XR, HIRSCH M, STAUDT A, KLAPDORKLEINGROTHAUS HV, CHING CR, HO TH 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 20(4) pp. 453-460 1993 

MATRIX-ELEMENTS FOR 0 NU-BETA-BETA DECAY CALCULATED WITH THE OPERATOR EXPANSION METHOD AND QRPA WAVE-FUNCTIONS 
HIRSCH M, WU XR, KLAPDORKLEINGROTHAUS HV, CHING CR, HO TH 
ZEITSCHRIFT FUR PHYSIK A-HADRONS AND NUCLEI , v. 345(2) pp. 163-169 APR 1993 

POSSIBLE SUPERNOVA MULTIEXPLOSIONS AND INDUCED GRAVITATIONAL COLLAPSE 
CHING CR, HO TH, WANG YR, YAO J 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 17(4) pp. 473-484 1992 

2-NEUTRINO DOUBLE BETA-DECAY WITH OPERATOR EXPANSION METHOD 
WU XR, STAUDT A, KLAPDORKLEINGROTHAUS HV, CHING CR, HO TH 
PHYSICS LETTERS B , v. 272(3-4) pp. 169-172 DEC 5, 1991 

A SEARCH FOR NEUTRINOLESS DOUBLE-BETA-DECAY OF CA-48 
YOU K, ZHU YC, LU JG, SUN HS, TIAN WH, ZHAO WH, ZHENG ZP, YE MH, CHING CR, HO TH, CUI FZ, YU CJ, JIANG GJ 
PHYSICS LETTERS B , v. 265(1-2) pp. 53-56 1991 

PION DOUBLE-CHARGE-EXCHANGE ON CALCIUM ISOTOPES 
CHING CR, HO TH, ZOU BS 
JOURNAL OF PHYSICS G-NUCLEAR AND PARTICLE PHYSICS , v. 17(8) pp. 1203-1208 1991 

Contribution of highly excited intermediate states in the pion absorption-emission mechanism of double charge exchange
Ching Chengrui, Ho Tsohsiu, Zou Bingsong
Nuclear Physics A, Volume 513, Issues 3C4, 9C16 July 1990, Pages 697-704

Contribution of the two-nucleon pion absorption-emission mechanism to the pion-nucleus double-charge-exchange reaction
Ching Chengrui, Ho Tsohsiu, Bingsong Zou
Nuclear Physics A, Volume 510, Issue 4, 23 April 1990, Pages 630-640

NUCLEAR FERMI MOTION EFFECT ON PION DOUBLE CHARGE-EXCHANGE 
CHING CR, HO TH, ZOU BS, CHIANG HC 
PHYSICS LETTERS B , v. 252(2) pp. 192-197 1990 

PRODUCTION OF RIGHT-HANDED GAUGE BOSONS AND HEAVY MAJORANA NEUTRINOS AT THE SUPERCONDUCTING SUPER COLLIDER 
HO TH, CHING CR, TAO ZJ 
PHYSICAL REVIEW D , v. 42(7) pp. 2265-2273 1990 

PION INDUCED DOUBLE CHARGE-EXCHANGE REACTION AT PURE NUCLEONIC LEVEL 
CHING CR, HO TH, ZOU BS 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 13(4) pp. 449-456 1990 

LONG-RANGE EFFECTS IN K0-KBAR0 MIXING CALCULATED IN THE POTENTIAL MODEL 
HO TH, CHING CR, LI XQ, TAO ZJ 
PHYSICAL REVIEW D , v. 42(1) pp. 112-117 1990 

NEUTRINOLESS DOUBLE BETA-DECAY : A NEW FORMALISM 
CHING CR, HO TH, WU XR 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 12(2) pp. 167-178 1989 

A NEW CORRECTION TERM IN NUCLEAR NEUTRINOLESS DOUBLE BETA-DECAY 
CHING CR, HO TH 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 11(4) pp. 505-508 1989 

A DISCUSSION OF THE QUESTION IN CONNECTION WITH THE EXPANSION OF A DIVERGENT SERIES ENCOUNTERED IN CALCULATING NUCLEAR 2-UPSILON-2-BETA DECAY MATRIX ELEMENT 
CHING CR, HO TH 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 11(4) pp. 495-497 1989 

A NEW METHOD FOR CALCULATING THE DOUBLE BETA-DECAY PROBABILITY 
CHING CR, HO TH 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 11(4) pp. 433-440 1989 

OPERATOR EXPANSION METHOD AND THE DOUBLE BETA-DECAY OF CA-48 
CHING CR, HO TH, WU XR 
PHYSICAL REVIEW C-NUCLEAR PHYSICS , v. 40(1) pp. 304-313 1989 

A NEW MECHANISM FOR LOW-ENERGY PION INDUCED DOUBLE CHARGE-EXCHANGE REACTION 
CHING CR, HO TH, ZOU BS, JOHNSON MB 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 11(2) pp. 171-179 1989 

AN ALTERNATIVE APPROACH IN NUCLEAR DOUBLE BETA-DECAY THEORY 
CHING CR, HO TH 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 10(1) pp. 45-51 1988 

THE RELATIVISTIC EQUAL-TIME EQUATION AND THE POTENTIAL MODEL OF THE MESONS SPECTRUM 
HO TH, ZHANG JZ 
ACTA MATHEMATICA SCIENTIA , v. 7(2) pp. 133-138 1987 

ON THE REACTION-MECHANISM AND THE SELECTION RULE IN THE DOUBLE CHARGE-EXCHANGE REACTION OF PIONS ON NUCLEI (DCX) : A STUDY OF THE CONNECTION BETWEEN THE 2 BETA-DECAY AND DCX ON JP=0+ NUCLEI 
CHING CR, HO TH, ZHAO WQ 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 6(2) pp. 187-194 1986 

THE MIXING EFFECT OF Z-DEGREES AND TOPONIUM 
HO TH, LIU JL, ZHANG XF, ZHU ZY 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 4(6) pp. 905-909 1985 

SOME DISCUSSIONS OF POSSIBLE MECHANISM OF Z0-]L+L-GAMMA 
GAO CS, HO TH, YE J 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 4(2) pp. 209-217 1985 

HOW TO DETERMINE THE NUCLEAR MATRIX ELEMENT IN 2-BETA-DECAY : A DISCUSSION OF PCAC, 2-BETA-DECAY AND DOUBLE-CHARGE EXCHANGE-REACTION OF PIONS ON NUCLEI 
CHING CR, HO TH 
COMMUNICATIONS IN THEORETICAL PHYSICS , v. 4(1) pp. 51-56 1985 

ON THE DETERMINATION OF NEUTRINO MASS : A CRITICAL STATUS-REPORT 
CHING CR, HO TH 
PHYSICS REPORTS-REVIEW SECTION OF PHYSICS LETTERS , v. 112(1) pp. 1-51 1984 

ON (PI-MU) ATOMS PRODUCED IN K(L)O DECAY 
CHING CR, HO TH, CHANG CH 
PHYSICS LETTERS B , v. 98(6) pp. 456-460 1981 

EXPERIMENTAL SUGGESTION TO TEST PROBLEM OF HIDDEN PARAMETERS IN QUANTUM-THEORY 
HO TH 
SCIENTIA SINICA , v. 20(6) pp. 740-741 1977 

YANGS SPHERICALLY SYMMETRIC, STATIC GRAVITATIONAL-FIELD 
CHEN S, HO TH, KUO HY, CHOU CL 
SCIENTIA SINICA , v. 19(2) pp. 199-206 1976 

POSSIBLE NEW PICTURE TO EXPLAIN EXPERIMENTS ON NEW PARTICLES : 2 NEUTRAL VECTOR GLUON MODEL 
HO TH, TAO H, WU YS, DAI YB 
SCIENTIA SINICA , v. 19(3) pp. 347-350 1976 

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

ӢѧѧDavid Willetts·ӵĺ

𾴵ķ

ף2012Լ˹ýЩͶ벢ðո֪
ѧʿ

Ҷίǵİ佱ʣҪһԼȵ潲ʵʱ
ֳҸ˵ľ⡣

¼ӢEdward Bulwer LyttonĻ

ʱȽ󡣿
ʦħȣԭ΢

ѧӦܹûп־ƫ̻ضΪǻѧĶ
ϣӢҹķ̰ϲѹ˿ѧϵĲͬ

ֿģ
David WillettsԱ

20121228

ԭģ

The Rt Hon David Willetts MP
Minister for Universities and Science

28 December 2012

Dear Mr. Shi Min Fang

Congratulations on being the recipient of the 2012 John Maddox Award 
in honour of those who devote passion and take risks to tell the truth 
about science.

I have read the Award Judges' citation and wanted to add my personal 
tribute to the courage you have shown in speaking truth to vested 
interest.

Your story reminds me of words by an English writer Edward Bulwer 
Lytton:

"The pen is mightier than the sword. Behold
The arch-enchanter's wand! - itself a nothing!"

Scientists should be able to pronounce on what they see as bad science 
without fear or favour and I am keen that in the UK expensive and 
disproportionate libel actions do not silence dissenting voices on 
points of science.

Yours sincerely

The Rt Hon David Willetts MP

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

̸ӷ̸۳ʱ

·̸ʵ¼ֻǴż¼뿴Ƶ

http://v.youku.com/v_show/id_XNTAwODEzMTg4.html

ˡǴҺãӭտڵĿ̸Ŀ
Сף뵽ˡαѧʿʦʦá

ӡãѴҺá

ˡڽĽĿУʦͬһ̽Ϊ
Ž𽥼٣ڽ϶𽥽͵󡣷ʦ
յĿǰŽʲôģ

ӡ˵ڽǺŨģΪһֱһڽ̵
ͳЩҵУⷽǱȽ϶صġŷޣ
̵ѾˣڱŷŽ̵ռ70%80%̫һ
Ž̵һֱȽ϶࣬80%ϵ˻ڽ̵ġҸյѧ
ʱ򣬾ͻ̡ɷǳ࣬ЩҪҰȥ̣
ǹ涨һҪ̣Ȼпá

ˡӲָꣿ

ӡԡһˣĸˣͲϵ͡
⼸ĿԿƣѾ˱仯ˣȻĿǰ˵
ռıǱȽ϶࣬ǳ˼仯һߣǸڹ
ԼԽԽˡǰһܼɻĴʣһ
˲˵Լߡ

ˡ˵

ӡԣΪһ˵߾൱ڲ£й
ѧйѧܶ඼۵ģϵѧ˵˼
̣㲻Ҫ˵Լۣ˼һú۹⿴㣬
ǲǵ°ܻҪ˵ǷͽǵͽҲ
ЩôǾˡ

ˡڽһ£

ӡԡҲһҾ˵ۣȻǾ͸ۡ
˵Ƿ̡̣̣Ǿһ㲻̫Ըˡǣ
꣬ͳһ˶˶˶һ־ǹ
վ˵ԼߣһЩҲ˵ǰ˵Լ
ߵĻǺٺٵġ

һ㶼˵һӰԱȣݽ̸ġһ
ӳˣȻҪȥ֤漰һıɱҪͥڷ
֤֤ͥͥ˵ĻҪǸʥģȻȾ˵
ߣҾͲʥ˵Ļ۴󻩣ɼǸʱ۵
Ҷǲ˵ġڲһںö˶վ˵Լ
ߣȶǴԼ˵ҲǰǺģ
һص㡣

ڶصЩųԼģڽ̵ˣҲˣ
ƽʱϽˡ

ˡ˵֯Ľ̻μˣ

ӡԡԾͳһΪҪоģ
ôһҪά԰ɣйԱ

ˡɱͦߵġ

ӡԣɱȷʵǺܸߣͳһЩþά
ֲȥˣάֲȥͰѽøˣáȥкüҽö
ˣĳ˹Ԣýϵı˵ĳɹԢԺܶ˻Ըȥס
ΪסǸرɡ

Ȼһƾ۵ˣҶö۵
˳飬ܹᣬҲǷǳġǰ
ļûУҿһͳƣϸʮǰͼ
ֻôʮйصд󲿷ֻۡ
⼸˲ȽϳľкüȻļר
дģңЩ۵˾ͳǣ
˵ʿĸر

ˡĸˡ

ӡ˵ʿʵǴӻ̵ʥģ̵
ʾ¼ĩˣħ֡

ˡļ

ӡļԣĸʿ

ˡ˵һġ

ӡģЩ˾ͺܽģǰڰʮ
ʮαѧʱ򣬵ʱαѧҲ˵ˣĸĴˣ
ڹԶӡꡢ˾ϣ˵ĴˣǾͺܽ˵
ǾĴˡȻʿҲһ֣˵ʿҲ
ʿ⼸˵оһѧҽ¡˹һ
ѧҽеأһǰǸѧɽķ˹
һҽп˹иϣ˹

ǳĸйص鶼ʮϰᣬ˶Ǵ03
04꿪ʼġ2001һܴ£9.11¼ֲܵϮ
¾ڶڽ̣ܴķ죬ʱܶ˶
ôһʣýҲôһ⣬ôֲ¼Ϊʲ
ôϵ۲ֹϵ۵ĻΪʲôȥֹôһ⣬
˺ܶ˵ַ˼ڽ̣۵ĿһЩ˾ͷ˱仯

911ĿֲϮ˹ԭּйϵڽйϵЩ
źܿȵԭּڽпֲϮģҲʹܶ˼˵
ڽ̲ǰ˵ǹϵģǺõģҲжһ档
ԾһЩǶڽ̡

2004ʱԭѧоɽķ˹һ顶
սᡷⱾͱΪ˶ĵһ飬ʱǳ䶯
˳а

˹һСϵ۵Ĵڰɡϵ۵˼
Ҳˣ˹ʮʱһС˽Ļ򡷣ܶ
˶רְȥˡǰǸѧңⱾϵ۵Ĵ
ȫǽ۵ģȫ˵ϵǲڵġԾôһȻԽԽ

ˡ˵2004꿪ʼ˶

ӡ˶

ˡô֮ǰʲôʽ˶˵֮ǰһ
߻ǲ֪ߣǺ֮ǰʲô
ϵأ

ӡʵ˶һֱУҪ׷ԴĻԷ
ֹϣѧͷģҲģǸʱһֱеģȻ
ոˣ˶Ȼ˶ˡ˶
ǰ˶ʲôΪʲôҪһأһ
ţǰ˶е˵ԼۣеĲ˵Լۣ
е˵ǲ֪ۣҰۣνĲ֪۾˵ϵ۴治
Ҳ֪ûϵ۴ڵ֤ݣϵ۵Ĵڣǲȥ֤ʵ
Ҳһʽۡ

ˡǲϵ۴ڣϵ۲ڵ

ӡԣһ𣬾ϵ۵ĴڣֻҪΪ
˵û֤ݣϵ۴ڵ֤ûеģҲžˣҪ֤ģҪ
ϵ۲ڣôҪ֤ģôһνĲ֪ۣ
ʵҲһֻۣǱȽۡǸղ˵ϵ۲
ǱȽǿģȽǿƵۡ

һЩ˵Ȼڷۡ޴ڣȻʵ
Ҳĳʽۣ˸񻯵һ˵ϵۣһ
һģȻԷʩģıģΪȻ
ɾ֣Ȼ֣ʹȻһģʵϾ
ĳ̶ֳϷĴˣ˵ʲôĻǲ
ʲô

˹ŵɯʼ۵ѧĳ̶ֳҲۣǱȽ
ۣǸʱ򹫿˵Լ۵Ļܻܵλ
̵ȺԾͻһơ

е˵Ĵڣֻڴʱһ¹
Ժ׿粻ˣȻĹԼȥУ
ٸԤ磬ԤȻ磬Ȼֻ
ԴǸط˸λã˵񲻴ˣˣʵ
ĳ̶ֳҲһۣΪûκιϵˣǰ
иָʽģڹ˵Լ۵˵Ƚ١

ˡٵġ

ӡøָ˵֪ѽѽ˵Լ
˼ߣΪȥܺ˵Լ˱ߣ˵ߣ
˼˵Ҳڽ̣ŵˣԾͽ˱壬ֻиָ
ģлߣҲУָ˵

µ۴ľ۵ţҲҪ˵ԼΪ˱
˵ӣΪ˱˵ȺҪԼһ£ǣҾ
ߣҲңҲȺһֺܹ
飬һҪص㡣

Ȼһص㣬ǰΪ˵۵ֻкܸ߲
ˣܹܺýˣ˵ͨĴڣȽĿҪڽ̵İο
ԾûбҪȥǽۡһЩ˼ôģ
Ҳ뷨ʱҹ
֧ۣǰһֱʾԼǲ֪ۣ
ۣΪڻ¼ͷѹϵ۴ڵЩ֤һһˣ
л̵ġԸ˵Լۣʱ
Ȱ˵ͨ۵ʱ졣

ˡǣҲһ˵˵Ǵģ˹̹
ܶŽˣһҥ

ӡǸҥԡ

ˡҥԡ

ӡ˵ڣǶҥԣĵǲġ

ˡߣ

ӡԡͱߣǲ֪ߣ˹̹
ˣ˹̹Ҳд˵ĳ̶ֳ˵ԼߣԳ
ǴȻ˹ŵɯϵۣҸղ˵˹ŵɯϵ۾ǴȻ

ˡȻ

ӡȻۣǷۡ

ˡ

ӡȻȻ˹̹˵ڽ̵˵
ԣϵͳĻԡǰ껹˰˹̹һš˹̹
񣬵ǰ˹̹һ㲻˵Լߣһַ۵ĸо
԰Ҳ˵Լǲ֪ۣôΰڵʱĻ
Ը⹫˵ԼۣôھͲһںܶ˶ĳ˵
ߣһҪ

ǰΪ˵ԼһֱȽϸߵεģΪͨĴڻ
Ӧһڽṩ֧ûҪȥۡڲһ
˶ͨҲܹ۵Ծͳ
һ˶Ϊʲô˶Ϊһ˶ҪռҪôҶ֪
ܣĻ۵ͻڸͳƶ⼸
ţ۵һֱģԽԽģڴ
ٷ֮ʮˣռ˿ġ

ˡܲ˵˶ҲŽ͵һԭ

ӡԣΪ۵Ļǿ϶Ͳȥڽ̡
˶һص㣬ǰۣߣҲдһЩ
ǶѧԵģдɳĺ٣ɳûУ˶
ܶ඼ǳ飬һ

һصǰ˶֤񲻴ڣҪǴѧ
ǴӿѧΪ˵ѧǸûйϵ
һֺе˵ѧû֤Ĵڻǲڣǿѧڽ
ͬ򡣹ϵ۴治⣬һѧ⣬Կѧ
ǲȥо治ڣҲû֤ϵ۲ڣҲû֤ϵ۴ڡ
˵ǰܶ۵ҲܡҪ֤˵ϵ۲ڣ
ǶǴѧϣ߼֤ġ

˶һ־ԿѧΪģ˹Ҳã˹
Ҳãд鶼ͼҪӿѧ֤˵ϵǲڵģϵ
ļǿԼһ裬ȻĽΪǲڵġ
ѧоһȻ֤ݣ֤ģô
ܾͳˣǷ֤ݣôƷˣ
˵һϵǷڵôһ裬м飬Ľ˵û
֤ݣ֤ǷģôҾͰƷˣȻǴӿѧ
Ƕо۵⣬ṩ˺֤ܶݡ

˹һѧңǰоǽۣѧңԵ
˹ǴĽǶо⣬оڽ⡣
ѧңʵһѧңҲо۵ģҲǴӽ
Ƕо⣬дһҲǱɳڽΪȻ
󡷣ǰڽΪһȻоôʵҲǴӽ۵Ľ
оڽΪʲô֣ڽֵ̳ԭԴΪ
гֵһȻǾ˵ϵǸڵģ˴ģ
ΪҪĳʱڽ̵ʱһƣ
ʷչ̵УںԭʼĽ׶ΣôһơȻʹǶ
֤˵Ȼγɵ󣬲˵һϵ
ôһ

˹ǰоѧѧĽǶ⣬
˵Ϊʲôе¸Уǰĺܶ˾˵ѧûģ
Щڽ̵˾Ϊ¸о˵ϵ۸ֵ¸У
ǾܹѧĽǶ˵ô£
ģģһյģĶǴϽġ

ҲԴӽ۵ĽǶоµԴ˵ںɵУҲ
µĳΣ˵ǽģ˵ϵģϵ۴
ģǴӸǶȣӿѧĽǶ֤ڽһȻ󣬲˵һ
ζϵĶһĶͿ԰ϵ۵ĸƷˣ˵
ȫġ

ˡղ˵˶ļݶѧйϵ

ӡԡ

ˡ֪Ҳѧʿǲѧڽ̣кǿ
ǲѧѧѧĿѧڽ̵ر٣

ӡԣû

ˡǡ

ӡûôƣܵĽԽߣôŽ̵ıԽ͡
ҲͳˣҸղ˵һĴ󲿷˶Ž̵ģģ80%
ģڿѧҵľֻ40%40%
ܾͿͷ˵ԼʵϿδδغϵģΪ
ͳƳģʹģ40%ͨĹ80%һԱȵĻ
ֻһˣһ룬ܳĿѧңѧ⿴ĸˣ
ѧԺԺʿΪĴѧңܳĿѧҵУ˼ûУ
7%ѧԺԺʿıֻ7%

ˡǳ١

ӡԣٺ٣ϸ7%ѧԺԺʿ
Щǵרҵᷢһ˼󣬾˵ѧҵУ
ѧԺԺʿѧҵıߵģǰٷ֮ʮȻĴ
ٷ֮ߡˡٵѧѧֻаٷ֮廹о
ˣ˵Ϊʲôѧıô٣ѧҵôࡣ
ѧоһȽϳ⣬ѧҶȻֿδرһ
٣ΪоĶѧĶܳġǿܶ
ȻôģôԴģЩǿܱԼҲϡ
Ϳģ˵ˣΪġ

ѧҲһΪѧ˽Ǳһ
Ҫ̵Ķ࣬һ˵Уȥǳǳĸӵṹ
ǿǺȻĽ͵ģۿԽ͵ģ˵ô
ôΪ֮ܶ񣬣ΪȻ̫ˣر
̫ˣȻγɵģӦϵ۴ģƵġ
ѧҿ˵ͨģҲҪôһ裬
ҺܶʵǺܲܵġһϵ۵Ļô
ƵģѧΪĿǱһ룬Ȼѧ
ԽãԽ룬ôԽȥΪʲôѧ
Уرǽܳѧҵмûԭ

ˡղҲᵽ˴ĵĽۣ֪ڴĵĽ
ЩʱһȺܶݶǽֹѧģ˺һ
ֹĽ۽ѧΥܵģôܲ˵ſѧ
չѧķչͿѧռô𽥶Կѧ
ˣԴĵĽ˽ĶˣӶڽһĵӶ
Ž½?

ӡԣΪһֱһһߡ
ܵĿѧĽԽ࣬ѧ֪ʶյԽ࣬ԴȻĿҲͻ
ۣŽ̵ֿԾԽ͡ڽ̽ʿʵҲ֪
ΪʲôǲȡָķҪƿѧرѧĽ
ôѧҪһǽ۵ĽΪѧоĻ
Ǹָѧ׷Դ׷Դȥ

ŷ޲һĵطŷ޶ߴûʲô
ģ۱ʼŷģһһֱһ
̴ͳŨĹңԭ̵ּһӪ

ˡ˵ԭּղҲᵽ911¼ԭּͽ

ӡԡ

ˡôԭּͽͨĽͽ˵ԭּɺ
ʲôͬ

ӡҾһ̵ӣ̵ĽɷǳĶ࣬
ԭּģзԭּģɵģɵģͽΪ˵
ʥĻǶ׼ȷģǾ䶼ģһʷļأ
кܶڹ£񻰣˵ԢԣôΪģǾΪ
ʥͷеûľͲҿԲأ
˵˵ںɵģɵġ

ʵĽͽд󲿷ֶһ֮࣬һ
ԭּĻͽȫߵģûĹ
ôߵԭּĻͽЩ˵Ϊʥÿһ仰ǶԵģ
ʥһ񻰣һ˵һʷļأǱÿһ
ѭʥ˵ϵ۴磬ôϵ۴磬˵ϵ۴ˣ
ô˾ϵ۴ģԭּ˵˵ɵĻͽʥ
ͷ˵ϵ۴磬ϵ۴ˣһִ˵һ£Ϊ
˵һһ£ʵ£ɵĻͽҲ
ţҲܽۡԭּĻͽͲţųۣȻʥ
˵ϵ۴ˣ۾Ǵˡ

ԭּϸͳģԺǾͿʼ
Խ۵Ľڶʮͳֹֽ̽۵
Ȼһѧʦȥۣ̽Ȼͱֹˣ
Ȼ߰Ʒˡһֱһʱ䣬رϷΪ
ϷĵطҲڽŨĵطԴҲԿ
һطԽĻŽ̵ҲԽ࣬ıĶ
ǾûȥЩ̽û⡣

ڽ۵ҪϷȻΪ
ʮǸȻĽǸѧʦûб
϶ǵºܶ˶Ըȥۣ̽£ôֱ
ʮ˱仯ʮһ£չ̫վ
һǷȥˣ¶̫ˣһֱԼ
һſƼǿ

ˡϴ

ӡ̫վȻǷ˼ĽΪѧ
УѧеԭΪڽ̵Ӱ죬Ȼʹʮ
ҶʼͿʼӿѧرѧ۵ĽȻһ
˿ʼЩֹ̽۵ķΥܷģ˾һֱ߷Ժ
߷ԺͲö˵ЩΥܵģĻѧУʼ̽ˡ

ȻЩڽ̵ǾһУòȥֹ̽ˣ
Ϊ߷ԺѾþˣǾ˵ֻһֿѧ˵ۣϵ
ģ˵ģҲһֿѧ˵ڽ̵ʱֶҪ
Ҫۣ̽ҪۣȻҪͬȵʱ̣νͬ
ʱ˶ʮʼġ

Ȼƽ̽ԣǾΪ˵һѧۣ
һڽۣǿѧģ㲻˵߷һ̣ĹѧУ
ǽ̵ֹģΪ̷룬ڹѧУԽ̽۵ǲ
ۡΪνƽʱ䣬ϷһЩͨˣҪ˵
ͬʱ̽ۺۣƽ̵ȥߣҲ˾һֱ
߷ԺȻöǷԺöһڽۣǿѧۣ
ڹѧУ̣ȡһʤȻֻܽ̽ˡ

ʱԭּǾһУ˵Ҳۣ
˵ϵ۴˵ƣ֮ô
һǻƣǻʲô˵

ˡۡ

ӡԣһܵĶܹ磬ǿ϶ֻ
ĵ˵رʣһЩϷݣΪЩȫ
ģϷһЩͨҪڽͷ˵Ҫͬʱۺ
ۣҲԣƽ̽Ҳȥ˾ĺüˣ
˾ʵҲûȫ꣬ÿδǶӮģΪһôȻ
Ũ񣬵ܷͷѾĹ涨ˣҲ׼ȷڽ̣
һ̷ԭֻҪõ˰˵ǮĽѧУ
ȥڽ̵̣ܴġ˽ѧУû˹㣬̻ѧУǿ
ġ

ԾΪôһԭҪ˾ĻӮģӮģֻ
ʱѡԾǸйأԿڽ̽ʿҲ֪
ѧڽһ޴вڿѧԽԽռ
̶ԽԽߣôҲǵŽ̵ٵģرǻԺ
ҪֿѧˣǰֻѧУôڿ
ԼȥȨĿѧվһѧУżһЩ
ʦǲЩѧģŵϵۡѧͿԼЩ
۵վЩѧվǾͿԼϣԼ֤ݣ
ʦţҲǺܹؼһȻһǣ⼸
ǽ̻˺ܶš

ˡǽ̻᱾Ҳˣ

ӡԡ̻⣬̻Уһǧ
̻һֱģҲνڽ̸ĸ˶ŷ޵ĵʱո
̵ģ·µڽ̸ĸҲΪ̻̫ܣҪԼŪһ
һֱġҶõִԺָܵԼԺ
ΪǰΪ̻⣬Ϊ̻Ȩ̺һ̻ͬʱ
ȨԺΪҿʼ̷룬˽̻
Ȩ

ô˵̻ѾһƵģȽϴ࣬ʵ
ûУ̻ʵһֱܺڰǰȻ̻ڲͨһЩ
ЩŸڸǣ⼸ֳŶˣְԱȥ
ͯԽڵĶַͯȵȣǴڳˣһ˽
ڲʵһֱ֪ĳһ񸸡ʦͯȻ
һȥЩȫˣȻԽ̻ᡢڽ̵
Ǻܴġ

һ˵뷨ЩְԱӦøϣڽ̵ĿӦñһ˸
룬ǵĵˮƽӦøߣ֤˷ǣĻŽ̾
ʲôã¶ĴҲǺܴġ

ˡղ˶пѧ̻
⣬ЩǵŽ͵ԭ

ӡԡ

ˡôˣΪһڽ̺ܷĹң
λһЩӰأ

ӡӰģϻӰ죬Ƕǽѡٵ
Ϊ󲿷ˣֱڴ󲿷ҲẒ̌ôЩ
Ͳ˵ԼẒ̌˵µѡ

ˡԡ

ӡŽ̵½ˣŽ̵ǲģ۵
ˣôͻһεĻЩҲã
ν磬Ͳܲǵ۵֧Ͳô޼ɵˣ˵
ǲǲµˣҲȥ֧ԭּ˶
ôĿŵˣͷ֧۵ˣǰ
ģСʲͳǹ֧Ҫͬʱ̽ۺ۵ġ

ˡģ˵ֻ˵

ӡпģΪСʲһԭּĻͽ

ˡȷģ

ӡԣϷݵģȻһνĻͽ
ʲô˼֮Ẓ̇ΪĸŴСţеĳ
ͲˣĻͽνĻͽأȻСţ
ǳԺҲɣᶨԼһν̣
ͽǾԼ˼оҪִСϡͿŵҪᶨĶ࣬
Сʲ֣ǴԺһν̣ȻֱȽ
ԭּġЩͳ°ǲ֧۵ģ
Ǻܼᶨģĺܶ漰ѧ⣬ʵ涼п
ѧڽ̵ִͷģһֱһ˵˵ѧڽǿԲ棬
ǿԵ͵ģʵǣԿ˵漰ϸо⣬
ⲻ˵ˣϸо⣬漰һѧڽ
⣬ѧ綼ΪϸӦоģ츣һڽ
ʿܲԸô˵ϸҲһ㲻ܰƻˣ
Ծ͵ϸоܵܶơǰСʲͳʱ
Ƹϸоģǰ°̨˾Ͳһˡ

漰һЩ⣬˵˹⣬˹
ϷģһЩڽʿһֱҪŪɷǷġһͬȨ⣬
ͬܲܽ飬ҾڽʿǱصģһֱԣ
мݶ˵ͬǺϷģеѾͶͶĽ
ͬǺϷģɾЧˣڽʿԵġͶ
ĳ̶ֳӳڽ˥ˣڶʮǰͶĻ
ͶȥģǸʱڽġ

ˡҲ˵ڽıԾ;µĻµĿѧ
ͷ

ӡԣڽеıصڽͷһЩȽϿŵģ
ȽɵģҲУǲϵģĳ̶ֳֻһֿͷ˵
ԼẒ̌Ĳˡ͵ܴڽ̵Щص
ĽѧĽǸģһֱַҵĽ谭


ˡҲ˵Ž˥ˣܶ˵Ļ
Ϳѧķչһõ飿

ӡԣҲһʶѧ磬ڶԿľ
Щڽ̵

ˡʦղ˵˰ˣôڻصй
Ŀǰйڽôģ

ӡĿǰйŽ̵ķһʱˡ

ˡʱУ

ӡʱˣܶ˶Լͽǽͽǰһϣ
ۣڽ̣ҿЩ΢
ܶﷴң˵ǱϴԡǰǸϴˣ˵
Ӧ֪ģҶģڽ̵ģйܶŽ̵رŻ
̵ģĳ̶ֳ϶Ӱ죬رĻ磬ѧ磬
ܶ˾ͱԼĻͽΪŻԺ󣬸ľͰ
ˣ˵йҪչĻҪΪ̻ũ
ҲúֺܿȵĴ̵ģЩصĽɣԭּĽ
ɣа̵Ľɡ

а̵ĽũﴫñȽ϶࣬ؽũҲУа̵Ľɱ
粻ǰνĩգ2012ĩµģȫɣǰж
磬ǸǴһа̽ɺݱģǴ

йĿǰẒ̌ʵڽܲɷֵģǽ
̸⣬ʵҲ漰й⣬ΪôĿǰһ
Ȼ漰һĻԵ⡣Ļкܶ෽棬
棬ƬģĻԡʵҪһܺ
˶ˣڽ̷ĻԣҸպǡǡڽͷ
صаһ֣ԡߴҵǣ㲻ҪŽ̵һ
ʱ֣ΪŽѾʱˣűʱ֣
ƣԽԽࡣԿң˵
ô˽֪˶Ž̵ģң;úܿЦ
ΪʵЩ˲˽֪ڽԽԽ˥
۵˶ԽԽǣҲΪʲôôڽܵйչһ
ԭ

ˡлʦľ˼˵ڽѻƬ
ʦһλӡر̣ڽ̸ľƣξƲմ
ȻܺãƷһҲ޷ͻġкþҲҾƣ
жƣ˵а̾ľơԣҪƷζþƣ̰
Զ붾ƣ԰֮ܶſѧȷĴ֮
ϧڡǽ̬ȡٴθлʦĿ
Ƿڽ̵ǰлѵտޣڿ
磬ټ

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

Ѻ΢̸̸жܷҩ

Ѻ΢20131923:30

С18΢  @ ʣ кָʲô

: ΣҪ棺άһ˲
תյ˥ߡֻ֢ѪҺ͸·Ƥ


Querida  @ ʣ òû˵˵ţ

: ý2003Ͳٱˡҽѧ
ϻܿ౨2007걱һҽԺ
ͯߣһһȴ

  @ ʣ ʹι涨

: ŷ޸ôձĴǡзȫ
ʮǰúҩͱƷҹ̨Ⱥ2003
2004ȫֹҩҩļƼ

̾޵  @ ʣ ᣿μǸʲôϵѽ

: Ƶֲձ麬ᣬеһ֡壬
мʮҩĺᣬڻõϸ١ѰǷ硢
ź⡢зͨɰͲܵȡ

95 @ :   ˪飩ҩҽưѪ
ӦҩֻҪѭҩԭ򣬾ûġϣû
ѧҽѧ˲Ҫ̰ѵõ֤ʵػùҽ
ѧ߽Ůʿסġ

: ǳȶĿǰûзֳʲôܹĶԣ
ԲҪѭνҩû⡣ʵҽҲΪ˺
ҩԼ˥ߵģ籱ҽԺһҽҵ
ҽ˶Ϊ˺ͬк˥ߡ⣬
ǰд֤һֻѧҩķֺҽûйϵ

  @ ʣ ˺ɢЩҩƷ᣿

: ǳ࣬ҩ¼гҩУҩĵм


wayh5΢  @ ʣ ɢУʲôҩƷʺֹȻ̵?

: ԱǲΪ˱Ϻͨһұ
һǲᳫֹȵġԹضҪƣҲӦԵ
򣬸Եأ翹ס̣Ǽ򵥵ǿֹȡ
СûгԹκֹҩȻʱΪðԡصӰ
Կʹֹҩõֹҩɳҡ

izCherise  @ ʣ ʹõõǲǾͲ
ˣ

: СطúҩֻùһΣͻ
ˣǲתġԲڡʹõõ⡣κζ
ҩʹöǲõġΪʲôԼ̨塢۶ȫ
кҩֻй½ûн

Amos  @ ʣ ҺӳʲôֹȰ

: ҼСʱΪðԣڻûгԹκֹҩ
˳Թ̩ŵֵҩҲûгԹκθðҩð֮󣬿Ҳὥ
ֹͣ

ѦСè @ ʣʦĿ£ѾҩˣҲ
Ȱ˵˲ҩܶгҩﺬҩʷʦôгҩ
ҩ?

: һҩƷ׼ĺţгҩע׼ĺǡҩ׼Z
ͷģZǱʾгҩҩѧҩõHгҩ
ȡúܺ顢ͨʲôģһ֪Ҫʲôġҩ
ǲ֪Ƶģɹ涨ҩƲЧҩɷ֣ɷ
вҩƣгҩġгҩڡӦһдšв
ȷҩϸ˵Ӧ

̵  @ ʣ Ϊҽ

: ¼ӳˣҽԼûȥҩᵼ
֢Զԣֻȫִҽѧȥ֡ӵǣִҽѧ
ҽҩĶԺҽٰϡƱ磬ͼжҩˡ

1998  @ ʣ ѵҽҩûһ

: ̸ҽҩ⣬̸ҽҩ⡣ҽҩ
ҽҩһģ޿ɱԡҩﲻӦ棬ҩ
Ҫÿѧϵͳо䲻Ӧнȫ˽⣬׼У
коûзֵ⣬Ҳȡȴʩҩ
ڷ滹ǸհףдǡӦвȷϡͿҲ
֪ôġ

ͷˮ @ :СʱٳԹҩ
ûٳԹҩˣ˹㶫ϲζʣȻ
˳˵һ䣺лΪᣬΪһ

: СʱԹҩʹڲˣɵȻڣ
Ϊǲתġǿܵ˲صĻ
ûӰ죬ֻҪֹܵһ˺㶫ҲԵģʱ
Ҳźҩɰ

taotaode΢ @ :Щ޷⣺ҳԣǸҳԣƫ
ƫҳԲ֪ʲôҩӳԣΪʲôأ

: ܶ˾ҩ˼ǧˣġʵûпѧݵ
ϡͿ˼ǧ꣬ͬ⣬һкܴ⡣ƽ
ʮ꣬ôܰǵҩ鵱أ

zgshh2002 @ :ʹйҽҩȨѵ֪
Σ𣿻ԭ

: йΣоǳ࣬ҩಿŵȻǳ
ΣҲҩΪάҽҩҵ棬
͸̨ȫֹҩ

332377  @ ʣ ʲôҩиð˱ĵֿܿ
ȥ

: ҩĸдС΢пв棬һŶ
ۡĸضҲתԱáҩ
ƵΡܶೣҩƵΣ绹ЩɰۻƵ
ҩ

ͷ׿ˮ @ : ڵйҽѧ磬ǲǺܶ
ųҩΪμСҡǶԴ˷أ

: йҽѧݬ룬ܶйҽµҽѧףѧ
רҵˮƽ籱ЭҽԺƾȻмҽҽƫ
ôЩҽ˽кûʲôֵġοкܶй
ҽΪҩؿۣ֪ҩкҲῪߣرЩҽ

ΰHiram  @ ʣ ԡԶô

: еĶĳּǱͨѧ֤֤
Ȩ֮ʹõġֻ˶β˲ģҪΪ
šԶҳԶҩԸΪƸð֮С


c4920  @ ʣ ĺӸðգôԸﳣҩ
Ʒʲô

: ȷǸðգһ㲻Сҩֻˮ
Ϣٴ·ٸǱɢȡյ÷ǳСеܣҾԵ
̩ŵա

ˮ  @ ʣ ǴҩƷʲô

: ҩƷͲӦ档ǴҩƷǺĿǰڵҩ
϶ٵģЩڰҽƭҩƭƭˣıƺ

СȺ @ :ӸðԳϢʹ˵ҩ
ˡ

: ϢʹӣƼΧ൱ȫģֻ
ʹʱŻࡢˣصĻᵼ¸˥ߣҪע
ҩһôעҲɲˡ

yztsf @ :ᱻѽ20꣬
й̨׷׶ԺֲƣйȻ˹ľͨ
ľҩֲԱҩƷ׼֮Сʹǽ
Ǽ֣ҲھӡƱý㷺עȲȡ
ʩ֮ӦSARS谷ش󹫹ȫҲ˷ǳƣ
Ҫҵ߷˿ѧ׼

: ڹѷΣʮ֮2003»
ع⣬ҩֲźܲԸؽҩû۵ѹ
ҩָԸȡжһʮȥˣӦüӴѹ
ʹҩʽӹ죬ҩ

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

΢רϷ԰ڣİȫ

Ѷ΢΢̸201311015:00-16:00

_Ұ : #ӡ˾İȫ# (@fangzhouzi) 籬ϣ
Գ20ûˮǼĳƷƿȪˮ֡Ӵô

 : ǷĸƷƿȪˮڻҿ϶Ϸĩ
ıһƪͨħˮˮġ

ŷ : #ӡ˾İȫ# @ λáиˮ
ˮʲô𣿶տ˺

 : һ˵ͬһطˮǸ
ˮӦúͷҵˮһտͿԺȡ

ϻ : #ӡ˾İȫ#ʷӣʳƷˮҩƷİȫ
⣬1000˿Ӱǲһ˿ܵܵĿʡ׷İȫ
УЩۿĳЩȨ˵ĲԴǷвȫУлл
ϣ˶ܲԻ

 : ˿ಢûо˰ȫС人ԺȥԹҪ
ǿִУܾνʡ׷ҵ۾߹٣ҵ
ˣ⼸Ҿ͵˼ðӵƻƷ
Ϸϵʳڶʥܡ԰ȫһֱҪעġ

С : #ӡ˾İȫ# (@fangzhouzi) ʦô
ѧҵҲͨпѰȫеģ

 : оӦΪҪѧоŻѧϰȤ
ΪҲȥУĿ̫Ͳȫ

 : #ӡ˾İȫ#ΪȫȫΪĹԱͬ
۵ûаȫأ

 : ڻа

ż : #ӡ˾İȫ# (@fangzhouzi) ѧٲ
ˣǷٵĺˡûĿˡ

 : ѧһֱոս¶人ѧѧԺԺФƽϮ
ġѧٵҪ˿վϣҲǾת΢


ϻ : #ӡ˾İȫ#ܶ˹˵ٵȴȫв
Ҿðвеط˵±ʧ˵ĵط
ûУĲȫô

 : ǰڲȥȥҲǳСġ̫
ȫҾӵȥ

Lanbo : #ӡ˾İȫ#йĻƶף˳
 ô

 : ָǴڵת󶹺װɡԼѾʮ
ˣ˵ͶӾˣĲȫҪ
á

 : #ӡ˾İȫ# ˾(@simanan) ƽһĿ
аȫ?ƽһѧƭ?

 : ûкƽһĿҲ֪ǸѧƭӡΪ
ħתʳƷԹ˺ܶԡҥԣƴѧڵ
ȫд½¶

 : #ӡ˾İȫ#˾ڻ360𣿷
ƾʲôԶԱ֤ݶʵģ

 : Աиоݣ360ֻҥҶ
Ա֤Ѿһ֤ʵΪʲôʵģ

Ŵ : #ӡ˾İȫ#ſѧѵͰȫ

 : ѧ֤ݺ߼ģŵĶͷ˵ſѧ
ìܵ˵ſѧȻܱ֤԰ȫܹӰȫĸ
ʡҽѧѧ㲻ҪҽҪûоҩͻ
ܵҩҩĶ

˭ : #ӡ˾İȫ#(@fangzhouzi) ̸İ
ɣһֱôȨåȫһ⣬
飬֧֣

 : лл֧֡ʽʦ͹ģҲ룬
֪ǣǰһܸҸթƭ浽˹
˰񣬹˰ȥ֤סƵܸ棬ø֪
˸ûĴڣյ˺ܶ

һV : #ӡ˾İȫ#˾ѲҪ
ˣڵаȫ𣿵ǵع𣿷
߹ٹǲаȫУ

 : ڴҥŮ׵åǲǾԼа
ȫУЩҥ̰ӵģҵʵݣӽһһ
ߣеѾԺ

С : #ӡ˾İȫ#֪λʦаȫ
ο˾ʦڵǩƬͦидġ

 : ˾ӦñаȫСҳڰȫĿǣ20109û
˹ˣȻǰΪ˻Һϲ˵һκһʿ
һĿȥģҳ˵ȥģС

huhuhu : #ӡ˾İȫ#˲ܳͷú˺ûаȫ
С۶Բԣ

 : ǵģʵҲȷˡҽ¶ô١ˣ
ֶûܵͷϮҵˣҲֻܵ΢ĳͷҺ
ûаȫС

 : #ӡ˾İȫ#ʷʦ˾ʦаȫ
е𣿰ȫأ

 : ûʲôȫСûаȫҲҪ֡һ
˵Ŀ꣬ôø˵һֵõġ

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

¶ȣߵ˵

ߣ

СϷһ֡˵硱һжǷģ˵
ָұߣ˵ζŴ˵ʾȵȡߵ˵粢
ǵҲšҿ˵ķʣ
ӦѰںܶʺΪ϶Ƿġ

Ԫչ¹ѧڶˡ˹ε±㷢һ³ɾͣʵ
˴ڱȾȻ͵ġ¶ȡ״̬塣ͨŽ籨
˶¶ȵĺ档ʵνġ¶ȡһ·Ҳǲ˼
ļ¡ǡǡ෴ǿ˵Ƿǳߵ¶ȣ޷ͨ¶ȸ
Ҳһ뾭෴ĵߵ硣

¶ȵʶʼճе飺ȡ䡣ɹ
𿾵ȿʹȡȵֿͨӴȷʽʹȣ
ͬʱԼ䡣¶ȱȳ̶ȵһ

ʼ¶ҲԴϤ״̬±ͻ
±궼18;ѷġǰ߽ˮ״̬Ϊ0ȣˮΪ100ȣ
ˮı˵Ȳա±ʾ¶ȶֻ
Եģֵû塣µġ¶ȡֻ˵ˮ¶
ҪͣڶܳûʲôѰĵط

19ڣѧͨѧоʶνʵз
ӻԭӵ˶ȵĴ˶Ĵݣ¶ȱǶԸ
¶Խߣ˶Խң¶Խͣ˶ڻɴ
ۣ¶Ȳû޵ġĳһ亮ĵ£е
ֹͣеԭӷӶֹ¶ȵļޣܴڱ
͵¶ȡ

1848꣬ӢѧҿľʿݴһΪѧ±ꡣνĿ
±ʵϾ±ֻ꣬±0ȡ±0ȱ
¶ȼޡҲǡȡ൱-273.15ȡˣˮı
ڿ±бΪ273.15ȣˮڵ¶ǿ373.15ȡ

ѧо֣ھ£¶ȣ״̬
ȱҲ޷ﵽġ˺ֵѧ֮ⲻ׼ԭ˵ԭǲ
Ծֹ˲ܴڴھȵϵͳĿǰ֪ӽȵ
ʵΪġѧֶͨοԽ״
̬ԭȴ£ʵֲɫ-˹̹ۡ2003꣬ʡѧ
Ժʵҽԭӽ450pK1pK10ĸ12ηĶȣڵ
¼¼

¶Ҳƽı־ͬ¶ȵŵһȵĻ䣬Ļ
ȣֱǶͬ¶Ϊֹ¶Ȳƽ̬Ψһ־¶
ͬһȾ˺ɫһȾ˻ɫˮӴҲụϣֱɫﵽһ
£ɫΪֹͬɫں͹һִĹ̡ǰ
ɫμϺһƬȣʧȥ˰ɫվӡ

ƽĹ̶νġ̡ͬ¶һ
Զشﵽͬ¶ȣȴԶػָһͷһͷ״̬ɫҺ
ԶϣȴԼصɫΡͬأһˮϣ
ˮȻɢȴ£صҲǳ˵ġˮա

Ϊֲ̣¹ѧ³򡤿˹1865
һءĸƧĴϣеԭǡתķ򡱡
˹ָһϵͳԷֵӵķݱ䣬෴Ĺ
ͨæʵ֡

ͳѧоΪ˸ΪĶ壺ֵϵͳڿ
ռе΢״̬ϵķ̶ֲȡһϵͳֻռСֵ״̬
зֻڹ̶ľ㸽񶯻߰ɫվöӵˮֵȽϵ͡
֮зӿȫ˶ͬɫںϺķӼķֲҲ
ӣֵҲͱȽϸߡ

ػΪ¶ȱṩһΪϸĶ塣Ϊ˶ϵͳΨһ
Դ¶ȼ򵥵ؿܵĺ׼ȷѧе¶Ǹıһ
ϵͳҪڲͬ״̬£һϵͳظıһʱҪ
ǲͬģϵͳ¶ȵĲͬ

ճУصı仯ǲһµģϵͳڻ
ͬʱػӡͣ״̬ռ䣬ӹ
Һ塢Ϊ壬ⶼĹ̡֮ʱ
Сó¶ֵ״̬仯ȻͬȴԶҲǾ
ϡ

ȻȴһЩϵճ
鲻෴ھӣϵͳӶܻԽԽ
ûֹռݵ̬ҲԽԽ࣬ϵͳػ
ӡ

еֻռӻ̬ӣԽԽ
ӻ̬ϵͳֹĸ̬ռ
Խĸ̬ϵͳԽߡ뾭ϵͳû𡣵ȷϵͳڸ
ͨþ

ڷǳ£ǿƳֻܼϵͳ
ϵͳУռݵ̬ޡӵĽʹԽԽ
ߵܼϡеĽϵͳطˡ
еӶܼϣϵͳȫ˱㡪
Ӷ̬ľϵľһֻȫߵˡ
Ϊӵؼ٣աıϵͳҪĶ壬ϵͳ
¶Ǹ

ϵĸ¶Ȼ˼ʵǺͱѧʶġ֮
ϡУΪھвܴڣҲҪǳ
ſܡĸ¶ϵͳ1951ͱѧںϵͳ
֤ʵˡͬʱѧҷ˼⡣ѡʵĲϺʹ
ԭֻܼɹԾǨȻԭӼе
ܼ̬ʹôڸ̬ԭӶڻ̬ԭϵ㴦ڸ¶
״̬ЩԭӲһµشӼ̬ԾǨػ̬ʱĹӱΪ

ͼϵͳǡ⡱ĸ¶ϵͳֻضɶ
ԭܼʵ˸¶ȣԭӱĻƽ¶Ȼ
¹ѧʵֵͻƱǰһЩԭͨ
ؽ˸¶״̬Щԭȫڸ¶ȣ¶Ȼ
ʵֵ״̬ǳȶֻܴǳݵʱ䡣

¶ϵͳӴ¶ϵͳǻᷢʲô󣿴ڸ¶״̬
ϵͳǲȶģԷͷͷŵı֡
Ӵ¶ϵͳʱԷؽݸԷ¶ϵͳ
ضӣ¶ߡͬʱ¶ϵͳʧʱû
ĻҲӣֱʧȥ¶״̬ϵͳѧҪ
ӵķݱ䡣ΪǴӸ¶һ
¶һ¶Ȳ¶ȸ䡱Ǳκ¶ȻҪȡ
һߵ˵硣

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

йУҪķʿƸĸ

ߣ

ǰйֻһȾĸ壬 
ʥ覵ĵʥأУҲȾߣҲƲᣬ
ӰǵĵʶΪ淶Ҷ˵ɳ˼Ӱ졣
ܵʽ廨ţӳϮģƾְȨȨĲ˽ձ
ѧԭУ͢20032011䣬ԼѧУںͻ
ְ֮㣬ȡ˻¸939ԪмԺܻд
͢ͽ̣ȨûղƲ100Ԫðϲ
˼Ӱ졣

 йУرѧԷӲʿΪѧ
ӢǵĲлŬѪս£ѧģټǹ
ѣ´ΣգΪάѧȨԣԶ
ǵжõ˰ڵ¼Ķ֧֡
ǱУ֮֮࣬жֻǱˮнУ
Ҫȫʦѧ룬Ĵ֧ԼӲʿ
ĴʿҪƶȽϽиĽľйУƶ
һЩԹ鿼ۡ

1 ѧİѧϢ֧ϢҪȫ񹫿йУչ
Ϊ˰ǮͶ룬Ӧ֪Ȩ ѧΪԪͶ룬
ǳÿ걨ңйز񱨸棬ʮϸ嵽ϸĿĿ֧

2ѧУ쵼ܼලȨСѧУԺֻ
ִȨûоȨشѧУ»ᣨ»ᣩߣ Ϊ
ѧУˣУԺֻʵѧУĹУȨȨ
ѧȨѧУ»ᣨ»ᣩ㱨Ȳȫʦ
ṫڵļලйУѧУУԺľȨִȨһ
ܷ֤ѵĺʹãЩѧԺίǲҪľڽѧ͵
ĹϣǲָѧԺľʹãְƵƸΣл
档ѧԺӦöȫڽʦйؾʹ ȫʦṫ
ڵļලרҽѧƽ衢ѧۡѧչѧУеҪ
á

3зִϡߴ·оܾ꣬йʮ
ᾭõķչΪ˴ίйزͨͽ෨
ɷ棬зУִϵ¾ƽУϢ
ǸУϢ˵2010 91ɽ䷢ġߵѧ
УϢ취ļơ취Ϊһ취
ȷᵽУԤ㡢ְסϽҽ
Ʊպᱣϻ漰ְش⣬Ӧͨ
ʽְ֪ Ȼźǣڸá취ִһ֮ʣй
ѧķġ2010-2011ȸУϢ۲챨桷ʾ
211 ̡е112ѧУûһѧУԴ
ȾԤ㷽ҲûһҸУʽľʹ
ĸֵ˼

4ʶУԺͷҪԡ㶫ʡίǻҫ
ȫʡУշعֳָУҪȷշ
ϾƣƽշعΪץ֣ȫƽУ
ͷܹйܲƣУܲơ
顣

ϣйУ䣬ٷ ˷͸ĸ漯
ŵţԭУʥһƬ

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

Ϻпѧѧоʱ

ߣȺ

Ϻпѧѧоΰ20113Ϻѧѧ滮
ϵпʱ鱨ְơΰ˲߱߻ְƣȴʱԼ
оԱȻÿʸ

ʮ塱ʱϺƼƻƸĸоϺѧ
ѧ滮ϵп⣬¼2011-2012꣬ĿΪϺпѧѧо
ΰ

ѧѧ滮

http://www.sh-popss.gov.cn/MoreInfo.asp?action=InsideNews&infoid=190

2011XAC004    ʮ塱ʱϺƼƻƸĸо
  ΰ    Ϻѧѧо

Ϻпѧѧоΰͷ٣Ҹжͬ걨
ʱ鱨ְơоԱ**յûãмְƻû2011
3ðְϺѧѧ滮ش⣬Ϻӿ콨衰
۳СоҺû¡

һоԱ**ֻ߱мְƣҲ20113ð丱ְ
б⣬ϺѯоĿչģʽ
սԹʾʾҲûܳɹ

ΰ鱨ְƣɹϺѧѧ滮⣬֪ʱ
ôѹصġͷ٣¸Ч£ֹ಻ε
Ϊ

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

ϺпѧѧоԼЪ˹

ߣϰ˼

ϺпѧѧоΰԼĲʿ٣̻λ٣
ԼЪ˹вżˡ

¼¼

ΰһҾǲ˵Ļᣡžžž״

ΰԶ飡ӣҸŷ⣬ĸΣ

ΰԱ֣㣨ˣû֣

ʿ֤ȷ䣬ĿҲ֤ɽȻΪּ
㹲Աΰȴʮ衪ⱳԵɣ
Ҫйȫ˼

Ϊʲôѧô꣬ΪȴԽԽⱣеβܹ
Ϊˡ

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

ƺӡ¡ʺذذӰ졷һƪϮ

ƺӡ༭

ųӻƽϾˮѧоԺˮˮԴоˮԴɳ
оҸΣڼ߼ʦ˶ʿʦһƪڡƺӡ
20125ϵ¡ʺذذӰ족һƪ
Ϯ¡ǶԹˮԴרҡܷгǰ
ϯʦˮSalman M.A. SalmanԭڹȨڿWater 
International3520104ϵȨġDownstream riparians can 
also harm upstream riparians: the concept of foreclosure of future 
usesĳϮԡϺͽṹϿڶԭĽм˵
͵ټ뼸Ĳο׶γɡ

ΪSalmanѣϣ󿯾ͿǳϮ
ԶųӻƽĳϮΪԹǴ𣬶ԱȨSalman
͡Water InternationalǸ



ͨѧѧԺ  ʤ

2013113

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

ƴšиڱƼѧƸУʧƫ

ߣɵ

    ˿.µϡ 13.01.08 ƴšиڱƼѧƸ
УһĵĲֹ۵ʧƫġ

    ԭָƼѧѴı䵽1984Ժ
׼԰оԺȫ22صУеΨһǡ985ѧΪ
̨УԽʮѧУĶ为вжΡ

    ƼѧЩ϶⡣ƬĽЩ
࣬ͬ϶ʧʣҲڱƴҵ֢ᣬҵ
ȷķչ򡣹ڡ985ѧǻûйϵ

    һְУʼ2004꣬985̵30ѧ292004
֮꣨ǰǩԼġ2004֮ǩԼֻһ 
    http://zh.wikipedia.org/wiki/985
    
    ڶƼѧЩķչܴ󻷾Ӱܴ󡣱ƴԭ
ұҵԺ1984ΪԺ׼԰оԺȫ22
صУ֮һұ𲿵֧Ϊؼ֮һұҵ1998곷
ƴûܹ빤ŲʧȥҪľ֧֮󣬱ƴ
ڹҽڲѧУǶƴǰ22º
ѡ԰оԺʱ򣬱ƴδܽ22ԺУ
ңн985̵Ĵѧ˱廪֮⣬
ʡĲ֧֡ƴñнԮĿҲ֧ܵ͡
ıʦ󡢱ŲŲȣȱƴʵھ
ԺУУ˱廪˴ֻũͱʦýĶ
֧֡Ѿ塢ʮҪĸؼ˲ţ
߿бڱƴҪ浶ǹغԺУѧϾ

    ѧУӦðҪʲôвͬġУ
ǲҪΪѧУ8δԺʿڵǰʱջ£ǲ˵8
УڣԷֲ͵ԺʿֻŪٵġݾ

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

ԡϾѧѧƸ־־ΪѧԵǸĳ

ߣ⽨

ã

Ϥ˿վһƪǰĵ
XYS20130104ΪϾѧѧƸ־־
ΪѧԵǸߡ˭˵Ǹ[1]
201012·ݵ˹(ARTIFICIALɣΣԣţ̣̣ɣǣţ
ã)־ϡ־⽨ε˳ϣ2ݺ[3йѡ
ͼ۲ֲԭΪ˵ǸעARTIFICIAL
ɣΣԣţ̣̣ɣǣţΣãӦΪArtificial Intelligenceָ˹
ܡڿΪAIJ

˭˿ųǵ¡ԡǡΪ˵Ǹ
Ǻʵزġ־ʱ䱦󣬶ϵ
ָزظΪ壬ѧйۣ˭
˵ΪԡǸΪбҪг塣Ҫ
ݰ㣺

һ㣺˭УAIJӢԭģӢԭ
еʵزṩһӢԭʵܽᣬ
ϸοӢԭġ

ʵǣ2010AIJһCorrigendum
˭ƵġԵǸšҪǣAIJĿۣ
ǣӢԭ[1]ߣΪûʵĵط
[3]Ǹǵ[1]Ҫףmain contributions
ԭoriginalġϽ۵õAIJ׼AIJ
ίԱһλⲿרҾϸԺóĽ
culmination⣬AIJࡢίרڴ˿
ѡôǡ塱This clarificationͬ[1]
νġѧԡ

ɴԶ׼AIJࡢίרΪ[1]Ҫ
ԭģҲ˵ڡ˭νġѧԡ
ͬʱ[1]֮AIJĿΪǸ Ҫ
2.1Сں͹ʽ2932[3]ӰҪ
ԭԡ

⣬˭Corrigendum/󡱷ΪǸšǷǳ
ߵġ

ǡ˭õĿӢԭģ2010꣺
Artificial Intelligence Journal,Volume 174,Issue 18,Page 1570
˿ĶĶжϡ

Corrigendum to Ensembling neural networks: Many could be better 
than all [Artificial Intelligence 137 (1C2) (2002) 239C263]

In 2002,we published in Artificial Intelligence an extension [1]
of a paper we presented at IJCAI-01 [2]. 

In Section 2 of the IJCAI-01 paper [2] and in Section 2.1 of the
AIJ paper [1],we presented a criterion for selecting a subset of
an ensemble of neural networks that could yield better performance
than using all members of the ensemble for regression.

The fundamental motivation for this criterion and its supporting 
details were ?rst presented in [3].Although we cited [3] on p.240 
of our article [1],we failed to do so as the source for Section 
2.1 and Eqs.(29)C(32) in Section 3,for which we apologize. The main
contributions of our paperthe subset search strategy (GASEN) 
introduced in Section 3 after Eqs.(29)C(32),the extension of the 
criterion to classi?cation in Section 2.2,and the empirical 
analysis in Sections 4 and 5are original. 

This clari?cation is the culmination of a thorough review of the 
papers [1C3] by the members of the AIJ Editorial Board and an 
expert external reviewer,and has been approved by the AIJ 
Editors-in-Chief.

References 
[1] Zhi-Hua Zhou,Jian-Xin Wu,Wei Tang,Ensembling neural networks: 
Many could be better than all,Arti?cial Intelligence 137 (1C2) 
(May 2002) 239C263. 

[2] Zhi-Hua Zhou,Jian-Xin Wu,Yuan Jiang,Shi-Fu Chen,Genetic algorithm 
based selective neural network ensemble,in: Proceedings of 17th 
International Joint Conference on Arti?cial Intelligence,vol.2,2001,
pp.797C802. 

[3] M.P.Perrone,L.N.Cooper,When networks disagree: Ensemble method 
for neural networks,in: R.J.Mammone (Ed.),Arti?cial Neural Networks 
for Speech and Vision,Chapman & Hall,New York,1993,pp.126C142.

Ϊ˷ߣṩ£Ľο

Ensembling neural networks: Many could be better than all 
[Artificial Intelligence 137 (1C2) (2002) 239C263]Ŀ

2002꣬˹ܣעָArtificial Intelligence Journal
AIJ˴ǰIJCAI-01עIJCAI˹ѧ
飩[2]չ[1]

 IJCAI-01[2]ĵ2ںAIJ[1]ĵ2.1Сڣһ
缯ѡһӼ׼ڻعܻñʹü
гԱõЧ

׼֧ϸڵĻ״η[3]
[1]ĵ240ҳ[3]ڵ2.1Сں͵3ڵĹʽ(29)(32)
û[3]ΪԴͷΪǸĵҪԭ
ģ3ڹʽ(29)(32)֮ӼѡԣGASEN2.2С
ڽòչ⣬4ں͵5ڵĻʵķ

˳AIJίԱһλⲿרҶ[1C3]꾡
ĽۣѾAIJ׼

עοӢԭͬ˴ʡԡ

ڶ㣺AIJĿ󷢱2010꣬ѾˡһҪ
[1]Ҫԭġţڴ¼ɣensemble
оԱ˵AIJίרҵۣǵ[1]Ҫ
ΪԭⲢܡȻĶ߿ҪһЩ
֪ʶĽܡһЩıͼϸڣϣܰԴ
Ȥ¼оĶжϣӶǰ˿
ж[1]һЩνġָء

µݱȽϳҲȽϼѾʹͨ׶
ӦܱⷽоԱ⡣

עݰʽŰϢĶPDF渽

ȸµӣ

[PC93] Perroneʿ1993£
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.32.3857&rep=rep1&type=pdf

[AI02] AIJ 2002£е[1] 
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.19.9955&rep=rep1&type=pdf

ΪʲôǵҪԭģ

[һ] Ҫݺͽ۲ͬ

1)	[PC93] оǡмɱʹõһá
	ժҪĵ4-5оȷָǷҪʾǡЧ
	ʹ硪Ҫκһ硱It efficiently uses 
	all the networks of a population C none of the networks need 
	be discarded˵ǶԻعѧϰоڵ
	2ں͵4ڷֱBEMGEMͼ
	мɣڵ3ں͵5ڷֱʵ֤

2)	оǡмɲѡһм
	ɡAIJµıȷָmany could be better than 
	allժҪ4ȷ˵ǵĹʾһֶȫ
	áit may be better to ensemble many instead of 
	all of the neural networks at handڹп˻ع顢
	棬ڵ3GASENѡһֽ
	ɣڵ4ڶԻع顢ʵ֤ڵ5ڽ
	bias-variance

3)	ԤѧϰͻعΣֻԻعоǲ
	óһԽ۵ġIJCAI-01Ļֻо˻عΣ
	AIJ 02ڿжԷͻع鶼оAIJ 02ڿ
	еóˡmany could be better than allһԵĽ
	ۡ[PC93]ֻԻعо

4)	ѧ˼[PC93]ϣ綼мɣǲ
	ϣмɡPerroneʿڷº1993
	׵NIPS93鷢һƪ£ĿǡPutting it all 
	togetherɴҲؿѧ˼ǲͬġ

[] еĹʽ

AIJ[1]ĵ2.1ڰ17ʽIJCAI2ڰ20ʽ 
IJCAI-01йʽ9-1018ȻɵãAIJʡˣ
ʽһ⻹AIJ µĹʽ29-32IJCAI-01йʽ21-
24Щʽ

1) ʽ1-13IJCAIʽ1-15ǳʶ֪ʶԵȵĶ壬
   Щʽм[PC93]еȼʽĹʽ֣ЩʶԵĹ
   ʽںܶ鼮жУұײõʹá˵

   a)	ʽ1-2ʲôǡȨֵʽ3-4ʲôǡȨƽЩ
	ǳ֪ʶ׵ںܶп磬[Markowitz,
	1952]p.78Ĺʽ[NNPR,1995]Ĺʽ9.95[KV,1995]Ĺʽ1
	[Rosen,1996]Ĺʽ4[SK,1996]Ĺʽ1[Wanas,2003] Ĺʽ3.8 
	Լ3.8֮ǰĹʽȵȡ

	[Markowitz,1952]: H.Markowitz,Portfolio selection,The Journal of 
	Finance,Volume 7,Issues 1,March 1952,Pages 77-91.
	http://onlinelibrary.wiley.com/doi/10.1111/j.1540-6261.1952.tb01525.x/pdf  
	(һƪ50ףGoogleScholarϱ18,600)

	[NNPR,1995]: C.Bishop,Neural Networks for Pattern Recognition,Oxford 
	University Press,1995.  
	http://nguyendangbinh.org/LyThuyetNhanDang/TaiLieuThamKhao/Neural%20Networks%20for%20Pattern%20Recognition.pdf  
	(ǱһǳĽ̿飬GoogleScholarϱ18,200 )
	
	[KV,1995]: A.Krogh and J.Vedelsby,Neural network ensembles,cross 
	validation,and active learning,NIPS1995.
	http://books.nips.cc/papers/files/nips07/0231.pdf   
	(Ǽѧϰ֮һGoogleScholarϱ1200 )

	[Rosen,1996]: B.Rosen,Ensemble learning using decorrelated neural
	networks,Connection Science,Volume 8,Issues 3-4,1996,pages 373-383.
 	http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.8.9813  
	(ǼѧϰҪ֮һGoogleScholarϱ200)
	
	[SK,1996]: P.Solich and A.Krogh,Learning with ensembles: How overfitting 
	can be useful,NIPS1996.
	http://books.nips.cc/papers/files/nips08/0190.pdf    
	(ǼѧϰҪ֮һGoogleScholarϱ200)
	
	[Wanas,2003]: N.Wanas,Feature based architecture for decision fusion,
	Phd Thesis,University of Waterloo,2003.
	https://pami.uwaterloo.ca:8443/pub/nwanas/thesis.pdf  
	һƪһѧĲʿģ

   b)	ʽ5-6ʲôһġʽ7-8ʲôһݼ
	ϵġҲǷǳ֪ʶ׵ںܶп
	 [NNPR,1995]Ĺʽ9.80[KV,1995] Ĺʽ3-479[Rosen,
	1996]Ĺʽ1[SK,1996]Ĺʽ2-3[Wanas,2003]Ĺʽ3.7ȵȡ

   c)	ʽ 9-11ʲôǡԡһ׵ںܶ
	п[Markowitz,1952]p.80Ĺʽ[NNPR,1995]Ĺʽ
	9.91[KV,1995]Ĺʽ13[Rosen,1996]Ĺʽ8[Wanas,2003]Ĺ
	ʽ3.9ȵȡ

   d)	ʽ12-13ǻڡԡԡдҲǻ֪ʶ
	׵ںܶп[NNPR,1995]Ĺʽ9.93-9.94[Wanas,
	2003] Ĺʽ3.10-3.11ȵȡֵָǣ֪д
	ݵ50 [Markowitz,1952; p.81]90
	ǳʶ֪ʶ[PC93]Ҳδõʹá

2)   ʽ14-16[PC93]δ֡

3)   ʽ17һʽǴʽ14-16Ƶó[PC93]ڵ6һ
     ʽڶʽڱκʽ17ƣָΪʽ
     ͬʽʵȫͬ

     ǵʽ17 
     [PC93]Ĺʽ 
    
    עϹʽTXT汾޷Բ鿴еPDF汾
     ַԭġ

   a)	ȣʽѧϲȼۡ˿ǵ˽
	Ƶ˵ʽǵȼ۵ģڡƵУȼ
	ǵN-1[PC93]ʽеNȻ˵ǵkN
	ЩԵģ磺ǵkһóN滻
	֮Ѿ˲ȼ۱仯ʵϣʽͬķ
	ϵͳдֱõʽӣһǵʽ17ڶ
	[PC93]ʽӣ׵ؿʽֻǡơѣѧ
	ȫȼۡ

	עϹʽTXT汾޷Բ鿴еPDF汾
	ַhttp://cs.nju.edu.cn/zhouzh/zhouzh.files/publication/emfa-ch6.pdf
	أ121ҳʽ6.3ʽ6.4

   b)   ڶʽȫͬһ˵N1,N2,,
	N20N3[PC93]ĹʽֻN3N1N2ԣN3N4,
	,N20ԣǵĹʽҪN3еN1,N2,N4,,N20
	ԡһ[PC93]ĹʽN3뼯ɺ󣬴˺ڿN4,
	,N20ʱN3ʼղᱻȥǵĹʽУڿN4,,N20ʱ
	N3Ȼܱȥֱͬо˼·Ĳͬ
	[PC93]һлһһ𽥼롱ķʽɼɣ
	һ룬ͻһֱ֣ǵĹǿȫϢ֮
	ѡԼɣÿһ綼ȥڡл롰ѡ
	Լɡǽֽһ˵

   c)	ֵעǣ [PC93]ʽǳConclusion֮ǰ
	ûгҪݵĵ25ڣλóֵСͨ
	Ϊdiscussion۶Ŀǰܵչϸ 
	[PC93] ͿԷ֣۽ܵĶ
	BEMGEMչˣ[PC93] еʽǰû
	ʽ֮Ҳûκ㷨ûж㷨ʵ֤
	ڼѧоУûƵûп㷨ûʵ
	֤вһͬһơ԰ȫļ
	Ǽûи۷Ҳûп㷨ûʵ
	֤ҲΪˡ硱

   d) 	ͬAIJ 2010Ŀ˵Perroneʿ˵һ
	Weighted AveragingȨƽ
	ܣͨÿѧϰһȨأȻмȨϣǸ
	һ㻯Ŀܣṩ˼ѧϰлعо
	跨ȥȷȨءͬȨȷʽ²ͬķͬ
	ۡ磬Ȩֵ趨Ϊ1/nõBaggingȨֵͨ
	Ҷ˹ȷõBayesian combinationȨֵͨ
	һѧϰѧϰõStacking˵[PC93]Ϊ
	ǵĹṩ˻跨ʹһЩȨΪ0͵õѡ
	ԼɡAIJίĵΪ[1]ȻѾ[PC93]
	صĵطٴ[PC93]ͬʱ
	ʵ[PC93]ʵ[1]ڱʲͬˣ[1]
	Ҫԭģνġԡ⡣

4)   ʽ29-32 ճӷ,һ֪Ĵ
     [PC93]ķʽ32 [NNPR,1995]Ȿ̿ʽ9.98
     AIµĵ246ҳʽ30ǰһУԼIJCAI01ʽ22ǰһУ 
     ȷ˵ճӷЩʽΪ˵ֱʹ
     ĴͳЧģǲҪԼķҲ
     ˵ʽ29-32ʵǲܱʹõġAIµʽ32֮Լ
     IJCAI01ʽ24֮ȷдIt seems that we can solve wopt
     from Eq.(32).But in fact,this equation rarely works well in real-
     world applications 

[] ǵǷ˼ԡ

νָءУһ˵ [PC93]һλΪλ
Ϊ[1]еġmany could be better than allwe can order the 
elements of the population according to increasing mean square error
 ...adding successively the ordered elements ...([PC93]12ҳ3)
if a network does not satisfy this criterion,we can swap it with the 
next untested network in the ordered sequence([PC93]12ҳ6ڵ
һ)Ե[PC93]˵ǡ𽥼롱Լ
ǽһδԹ罻

many could be better than allǳڼѧϰоУˣ
Ӧڼѧϰоﾳ½⣬ܼ򵥵ֱӴӢȻ
ĺȥ⡣ȻԵ⣬ΪֻҪûбʹãmany 
could be better than allһ⣬ԭ£

1)   ڼѧϰ򣬡лɼɣһһؼ뼯ɣ
һֳͨÿһʱҪĳּ飨sanity 
check罫루㷨Զֹͣ
FreundSchapireBoostingBoostingУ
ĳѧϰ0.5룩ù1990귢
[PC93]֮ǰ2003ġ¶д¼ѧϰо
˶Ϥࡰлл͡лǼѧϰ
෽ǰBoostingΪBaggingΪڼѧϰо
ﾳ£Ȼmany could be better than allǵݡ

2)   ǵġѡԼɡҪ֮һǴһall-member ensemble
ɵļɣнѡ񣬡many could be better than 
allǷӳһ㡣ŵǣӡallѡһ֣Ա
allá˲ͬǣ[PC93]лķǴӡall
ȷȥͨsanity checkĸܹ֮ȡallá
Уκθڡ롱֮󣬾Ͳٱȥֻ֡ķ
ѡԼκθ嶼ܱȥȻͬ

3)   ֵעǣ[PC93]λǳConclusion֮ǰһ
УûиƵûκ㷨ûж㷨֤ʵϣ
[PC93]ǵĿǰмɣֱӰ
ŵһڼлΪеƶֲ̬⣬ˣ
˵ܿԡԲⲡ̬⣬
ϣȰ簴Ȼ󰴾ȴӸߵͼ뼯У
̬磬ͺһ罻һֿǡԵ
лǵѡԼ¡

4)   һ˵ʹ [PC93]ʵǵġ롱Ҳǵġѡ
ԼɡȫͬǵġѡԼɡǼ򵥵ظݾѡǵ
ѾվѡǲõģѡԼ𾫶
ߵĶѡ񾫶ȵ͵磻ƵҲδһʱ
ôˣ˼·ϻڷϣǵĹ [PC93] Ȼͬ

[] СͬУͬоĹרңĿ

1)   [PC93]ǵ[1]ǡѧϰensemble learning
     ɹGoogleScholarʾǰѱ759Σѱ810Ρ
     кܶͬʱ˽ϴƪЩµ
     ߶ؿĹǲͬġ磺

    a)	Garcia-PedrajasڡIEEE Transactions on Evolutionary 
	Computation2005£http://cib.uco.es/documents/Garcia05TEVC.pdf
	У[PC93]ǣSeveral works have shown [Perrone&Cooper
	,93] that the network ensemble has a generalization error 
	generally smaller than that obtained with a single network 
	1ڣ1ΣǵĹǡSome recent works have 
	shown [Zhou et al.02] [19] that the combination of a subset of 
	all the trained networks can be better than the combination of 
	all the networks2ҳߵ3ΣԼ[Zhou et al.02] 
	have shown that a combination of some of the networks may be 
	better than a combination of all the networks,and that a genetic 
	algorithm can be used for obtaining that subset of networks.
	2ҳұߵ2Σ

    b)	RooneyڡIntelligent Data Analysis2006
	http://iospress.metapress.com/content/cny5uaf5n5l9cuvn/fulltext.pdf
	ж [PC93]ǡThe simplest ensemble method for regression 
	is referred to as the Basic Ensemble Method (BEM) 
	[Perrone&Cooper,93].BEM sets the weights i to be equal to 
	1/N.This method does not take into account the individual 
	performances of the base models ...The generalized ensemble 
	method (GEM) and Linear Regression (LR) were developed to 
	give more optimal weights to each base model.However,both GEM 
	and LR techniques may suffer from a numerical problem known as 
	the multi collinear problem.49ҳ4-5Σ
	ǹǣIt has been shown that given the presence 
	of N models it is possible that an ensemble learner can perform
	better if it only uses a given subset of those models rather 
	than all [Zhou et al.02]50ҳڣΣ

2)   ̿ [C.Bishop,Neural Networks for Pattern Recognition,
     Oxford University Press,1995] ǰΪ [NNPR,1995]ڵ
     9.6ڽ [PC-93] Ĺ. [PC-93] صҪ
     ʽڸýڿǰڶֵ˵ڸý
     ȫҲmany could be better than allƵ˵
     ɴ˿ԿͬרҲûΪ[PC93]ѡԼɵ˼
     롣ֵָǣ[NNPR,1995] ǷǳĽ̿飬ܶ
     УʹøΪ̲ģдĶߣGoogleScholaró
     18,200Σǡԡ[PC93]ĹȻܹ
     ӹרԼ˶ߡߵ۾

3)   ѧϰҪѧMCS (International Workshop on Multiple 
     Classifier SystemsһĻ飬conference
     workshop)־MCS20098MCS飩λ
     ֮һMCS20109MCS飩Panel speaker֮һ,
     λPanel Speakerģʽʶ߽K.S.FuHorst Bunke
     Robert DuinLucy KunchevaTerry Windeattȼѧϰ
     ѧߡָνԡĹ־ڸΪ֪
     Ĺ֮һлɼɵķ[PC93]ķһ֣
     ͨΪѧϰ෽֮һǵĹ[PC93]
     лɼɵķôǣ־ھ
     ȻΪ˺Panel Speaker⣬ǵġ2010
     ȨڿAIJ־2012׵ѡIEEE Fellow
     IAPR Fellow2013IEEEѧܳɾͽȫ
     ÿ䷢һλ40µĽܳѧߣɴҲɿ
     ͬרҲûΪǵĹԡ

֮ܶЩͬרAIJĿбAIJίʶһµģ
[1]Ҫԭģ[PC93]ͬȻҲڡԡ
ϵ

ж[PC93]гֵƹʽĳ˵пԿ
ظĹʽڼѧϰǳʶ֪ʶظҪʽ
1995̿[NNPR95]жܿʵϣ[PC93]Ҫ
ףWeighted AveragingȨƽܣÿ
ѧϰһȨأȻмȨϡ[PC93]ķܣ
ԼǵĹ־2012Chapman & Hall
ӢרEnsemble Methods: Foundations and Algorithmsϸ
ĲȤĶ߿Խһοе½ڣ

4.2.2ڣhttp://cs.nju.edu.cn/zhouzh/zhouzh.files/publication/emfa-sec4.2.2.pdf 

6.2ڣpp.121http://cs.nju.edu.cn/zhouzh/zhouzh.files/publication/emfa-ch6.pdf  

ǵArtificial IntelligenceǵԶԶֹѧϰ
ѧϰĶ߶ر֪ʶδǰʽĵطҪ
ãʿ⣬AIJ˸ÿ
Ϊ֮ǰһпĶճ
ӷʽ29-32[PC93]

λ߶Աһƪ£

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.32.3857&rep=rep1&type=pdf 

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.19.9955&rep=rep1&type=pdf 

˳ʶԱ֪ʶ[PC93]ֻĩβ12ҳһʽǵʽ17
ƣʵʲͬĿġ۷㷨ʵ鶼¡
ʱĶƪ˵ĶߣǷԡѾжϡ
ϣƪά˿Ŀ͹ۡԡ

201319

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

»ƵԸ

ߣŶ

ǰ ĳҽԺĴȾһλʮŮʿǻƵ
Ը׵סԺˣң ڳ˴ġҺ֣
ˣôô϶˵ҽʱһ˵ҳԹ
裬϶סҽ˵ΪԴƵԸ׵
ˣѲһˡ

ƵԸףڴȾҽǳʶź
ǣżδȡʩ֪ʵ⡣

ڳԴˣĸ໹ͦסô

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

360͵׬Ӷ𣿲Ա꿪ʼ

Ͽб20130111

360͵׬Ӷ
ʼʡ

-Ͽб111Ѷ 飩ʼԱڵ
ҳƻ360ȡҡԱ͡Ӷ𣬺һ
Աָͨ͡ԱƹƷճ
ЧӶĸ˻߼塣Աұʾ꿪ͨԱͣ
360ƷǷͨԱӣ궼׬ȡӶ

깫棺Ա360

¦һҵȼﵽ4Ůװ̣ҳϹ棺Ϊô
͵ļ۸򱦱ϣҲҪͨ360±Ϊƹ
360һֱڰпȡԱӶǵĳɱë
ϣΪ˸λ׵棬һ±

ɷУ˽⵽ĿǰʡšȪݡȵأҲʼ
360ҡΪһЩҾѡԼرԱ͹ܣ
ԱӼܡ

׬Ӷ1600 ̩滢кιϵ

ҪԱ600ͷءǼ׵£ǣԱͳΪ
ҵԱͨƹƷɽԼ
ȡӶ5%~30%ԽԽ꿪ʼԱͣΪǻɣԽ
Խռ360͵׬Ӷ𡣾ݳƣ360360Աͣûͨ
360ԱƷԶתӶı

ҵĽû360йƷʱ㲻ǴԱ͵
ӵ룬ҲܲӶһЩIDĿڴУЩIDžָˡϺ
̩Ա͡ʾϺ̩Աаӵڶݲ㣬һ
Աɸߴ1600Ԫҡ

Ѹ滢˹й˵ָϺ̩Ƽ޹
˾滢360˾ڹ̩ķ˴½滢360һλ߹ͬ
⣬Ϻ̩360100%عɹ˾޹˾źԼ
Эϵδõ֤ʵ

ϵ360360ʾԾԺʵʵϣԱ
ġɡ2012ֹʱ360ƣ360ȫٳԱ
Ӷ𡱾ΪҥԣԱͿɷʹ360ȫƣ360˾һ
û棬û360˾ͷչߡ

(XYS20130113)

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

һΪ˰޷˿վĹ˽˿վĶ̬ؿ
˿ʼбxys@yahoogroups.com

ʼбÿ췢һ˿վ¸µϣÿµ׷һÿ
Ŀ¼ʼбĶμ롰˿֮ѡ
ʼбxys_friends@yahoogroups.com

ù䶩ġù䣬ܿ޷յ
Ƽ䣺yahoo.com, hotmail.com, gmail.com

ġ˿ʼбĵ䷢һհż

xys-subscribe@yahoogroups.com 

塢ͣ˿ʼбͣ䷢һհż

xys-unsubscribe@yahoogroups.com 

Ͷxinyusi@yahoo.comôıļͶ塣

˿(www.xys.org)(xys7.dxiong.com)(xys.ebookdiy.com)(xys2.dropin.org)

