Journal of Universal Computer Science, vol. 6, no. 1 (2000), submitted: 20/10/99, accepted: 6/11/99, appeared: 28/1/00 Springer Pub. Co.

Size: px
Start display at page:

Download "Journal of Universal Computer Science, vol. 6, no. 1 (2000), submitted: 20/10/99, accepted: 6/11/99, appeared: 28/1/00 Springer Pub. Co."

Transcription

1 Journal of Univrsal Computr Sin, vol. 6, no. 1 (000), submittd: 0/10/99, aptd: 6/11/99, appard: 8/1/00 Springr Pub. Co. Galois Conntions and Data Mining 1 Dana Cristofor (Univrsity of Massahustts at Boston Dpartmnt of Mathmatis and Computr Sin Boston, Massahustts 015, USA dana@s.umb.du.) Laurntiu Cristofor (Univrsity of Massahustts at Boston Dpartmnt of Mathmatis and Computr Sin Boston, Massahustts 015, USA laur@s.umb.du.) Dan A. Simovii (Univrsity of Massahustts at Boston Dpartmnt of Mathmatis and Computr Sin Boston, Massahustts 015, USA dsim@s.umb.du.) Abstrat: W invstigat th appliation of Galois onntions to th idntiation of frqunt itm sts, a ntral problm in data mining. Starting from th notion of losur gnratd by a Galois onntion, w dn th notion of xtndd losur, and w us ths notions to improv th lassial Apriori algorithm. Our xprimntal study shows that in rtain situations, th algorithms that w dsrib outprform th Apriori algorithm. Also, ths algorithms sal up linarly. Ky Words: Galois onntion, losur, xtndd losur, support, frqunt st of itms Catgory: H..0, E.5 1 Intrution Galois onntions ar algbrai onstrutions whih play an important rol in latti thory, univrsal algbras and, mor rntly, in omputr sin (s []). W dmonstrat thir usfulnss as an algbrai and onptual tool in dsigning int algorithms for th idntiation of frqunt sts of itms, as dnd in data mining. Lt (P; ) and (Q; ) b two partially ordrd sts. A Galois onntion btwn P and Q is a pair of mappings (8; 9) suh that 8 : P 0! Q, 9 : Q 0! P and: for x; x 0 P and y; y 0 Q. x x 0 implis 8(x) 8(x 0 ); y y 0 implis 9(y) 9(y 0 ); x 9(8(x)) and y 8(9(y)); 1 C. S. Calud and G. Stfansu (ds.). Automata, Logi, and Computability. Spial issu ddiatd to Profssor Srgiu Rudanu Fstshrift.

2 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions It is asy to vrify (s []) that 8(9(8(x))) = x and 9(8(9(y))) = y for x X and y Y. Lt (P; ) b a omplt latti, that is, a post suh that for any K P thr xist both sup K and inf K. Alosur is a mapping l : P 0! P suh that th following onditions ar satisd: 1. x l(x),. l(x) =l(l(x)) and 3. if x x 0 thn l(x) l(x 0 ), for all x; x 0 P. An lmnt x P is l-losd if l(x) =x. Th st of l-losd lmnts will b dnotd by C l. For any Galois onntion =(8; 9) btwn th omplt lattis (P; ) and (Q; ) th mapping l = 98 is a losur on P whil th mapping l 0 = 89 is a losur on Q. It is asy to s that in this as th mapping = 8C l is a bijtion btwn C l and C 0 l. A standard mth for gnrating Galois onntions is through th notion of polarity. Lt X; Y b two sts and lt R X Y b a rlation. Dn th mappings 8 : P(X) 0! P(Y ) and 9 : P(Y ) 0! P(X) by for K X and 8(K) =fy j y Y;(x; y) R for all x Kg; 9(H) =fx j x X; (x; y) R for all y Hg; for H Y. Th pair =(8; 9) intrud abov is a Galois onntion and is usually rfrrd to as th polarity on X and Y dtrmind by th rlation R. Dnition 1.1 Lt =(8; 9) b apolarity on th sts X and Y. Th smidistans gnratd by ar th mappings d : P(X) P(X) 0! N and d 0 : P(Y ) P(Y ) 0! N dnd byd (U 0 ;U 1 ) = j8(u 0 ) 8 8(U 1 )j for U 0 ;U 1 P(X), whr 8 is th symmtri dirn opration, and d 0 (V 0 ;V 1 )= j9(v 0 ) 8 9(V 1 )j for V 0 ;V 1 Y. Th proximitis gnratd by ar th mappings p : P(X) P(X) 0! N and p 0 : P(Y )P(Y ) 0! N dnd byp (U 0 ;U 1 )=j8(u 0 )\8(U 1 )j for U 0 ;U 1 P(X), and p 0 (V 0 ;V 1 )=j9(v 0 ) \ 9(V 1 )j for V 0 ;V 1 Y. Th wight funtions ar th mappings w : P(X) 0! N and w 0 : P(Y ) 0! N givn by w (U) = j8(u)j and w (V ) = j9(v )j for vry U P(X) and V P(Y ). If th Galois onntion is lar from ontxt thn th subsript may b omittd. Also, if a st onsists of only on lmnt ` w may writ simply ` instad of f`g. Proposition 1. Lt b apolarity on th sts X and Y. W hav 1. d (U 0 ;U 0 [ U 1 )+d (U 1 ;U 0 [ U 1 )=d (U 0 ;U 1 ),. p (U 0 ;U 0 [ U 1 )=p (U 1 ;U 0 [ U 1 )=p (U 0 ;U 1 ), 3. d (U 0 ; l (U 0 ))=0, for vry U 0 ;U 1 P(X).

3 6 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions... Proof. Th dnition of d implis that whih imply d (U 0 ;U 0 [ U 1 )=j8(u 0 ) 8 8(U 0 [ U 1 )j = j8(u 0 ) 8 (8(U 0 ) \ 8(U 1 ))j = j8(u 0 ) 0 8(U 1 )j d (U 1 ;U 0 [ U 1 )=j8(u 1 ) 8 8(U 0 [ U 1 )j = j8(u 1 ) 8 (8(U 0 ) \ 8(U 1 ))j = j8(u 1 ) 0 8(U 0 )j; d (U 0 ;U 1 )=j8(u 0 ) 8 8(U 1 )j = j8(u 0 ) 0 8(U 1 )j + j8(u 1 ) 0 8(U 0 )j = d (U 0 ;U 0 [ U 1 )+d (U 1 ;U 0 [ U 1 ): For th sond part of th proposition w not that p (U 0 ;U 0 [ U 1 )=j8(u 0 ) \ 8(U 0 [ U 1 )j = j8(u 0 ) \ (8(U 0 ) \ 8(U 1 ))j = j8(u 0 ) \ 8(U 1 )j = p (U 0 ;U 1 ) for vry U 0 ;U 1 P(X). Th proof of p (U 1 ;U 0 [ U 1 )=p (U 0 ;U 1 )isntirly similar. Finally, not that d (U 0 ; l (U 0 )) = j8(u 0 ) 8 8(9(8(U 0 )))j = j8(u 0 ) 8 8(U 0 )j =0. Proposition 1.3 Th wight funtion w X and Y has th following proprtis: gnratd byapolarity =(8; 9) on 1. maxfw (U 0 );w (U 1 )gd (U 0 ;U 1 )+p (U 0 ;U 1 ),. p (U 0 ;U 1 )=w (U 0 [ U 1 ) minfw (U 0 );w (U 1 )g, 3. d (U 0 ;U 1 )+p (U 0 ;U 1 ) w (U 0 \ U 1 ),. w (U 0 )+w (U 1 )=d (U 0 ;U 1 )+p (U 0 ;U 1 ), 5. w (U 0 )=w (l (U 0 )), for vry U 0 ;U 1 P(X). Proof. W giv th argumnt for th third part of th proposition. Not that d (U 0 ;U 1 )+p (U 0 ;U 1 )=j8(u 0 ) 8 8(U 1 )j + j8(u 0 ) \ 8(U 1 )j = j8(u 0 ) [ 8(U 1 )j j8(u 0 \ U 1 )j = w (U 0 \ U 1 ): Th rst of th proof is lmntary and is lft to th radr. Proposition 1. Th proximity funtion p gnratd byapolarity =(8; 9) on X and Y has th following proprtis:

4 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions p (U; U) =w (U). p (U 0 ;U 1 )=p (U 1 ;U 0 ) 3. p (U 0 ;U 1 )=p (l (U 0 ); l (U 1 )). p (U 0 ;U ) p (U 0 ;U 1 )+p (U 1 ;U ) 0 w (U 1 ), for vry U; U 0 ;U 1 ;U P(X). Proof. Th argumnt is lmntary and is omittd. Proposition 1.5 Lt U 0 ;U 1 X and lt b apolarity on th sts X and Y. For vry sts U 0 ;U 00 suh that U 0 U 0 l (U 0 ) and U 1 U 00 l 0 (U 1 ) w hav d (U 0 ;U 00 )=d (l(u 0 ); l(u 1 )). Proof. Not that if U 0 U 0 l (U 0 ), thn 8(U 0 ) 8(U 0 ) 8(9(8(U 0 ))) = 8(U 0 ), so 8(U 0 )=8(U 0 ). This implis d (U 0 ;V 0 )=j8(u 0 ) 8 8(V 0 )j = j8(u 0 ) 8 8(U 1 )j = d (U 0 ;V 0 ). Frqunt Sts of Itms and Closurs of Sts of Itms Th notion of frqunt sts of itms is formulatd starting from two nit, nonmpty sts, a st of transations T and a st of itms I, and from a rlation R T I. Th mappings of th polarity dtrmind by th rlation R ar dnotd by ti R : P(T ) 0! P(I) and it R : P(I) 0! P(T ). If R is lar from ontxt, th subsript will b omittd. Th tabl assoiatd to this Galois onntion is a tabl whos hading is th st of itms I. Eah itm, rgardd as an attribut has th binary domain f0; 1g. Ift is a transation in T, thn T is rprsntd in th tabl by a tupl (dnotd by th sam lttr) suh that t[i] = 1 if and only if (t; i) R. Dnition.1 Lt b apolarity on th st of transations T and th st of itms I. Th support of a st of itms K, K I, is th numbr supp (K) = w (K)=jT j. A st K is -frqunt if supp (K) and is -maximal if it is -frqunt and thr isno-frqunt st L suh that K L. Th wight of a st of itms K is givn by w (K) =jti(k)j = jt T j t[k] = 1 for vry k Kj: In othr words, th wight of th st of itms K quals th numbr of transations that ar assoiatd with vry itm k K. In viw of Proposition 1.3 w hav Thorm. If isapolarity on th st of transations T and th st of itms I, thn for vry two sts of itms K 0 ;K 1 I w hav: 1. maxfsupp (K 0 ); supp (K 1 )g(d (K 0 ;K 1 )+p (K 0 ;K 1 ))=jt j;. supp (K 0 [ K 1 ) minfsupp (K 0 ); supp (K 1 )g; 3. d (K 0 ;K 1 )+p (K 0 ;K 1 ) supp (K 0 \ K 1 ) 1jT j;. supp (K 0 )+supp (K 1 )=(d (K 0 ;K 1 )+p (K 0 ;K 1 )) 1jT j; 5. supp (K 0 )=supp (l (K 0 ));

5 6 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions K 0 K 1 implis supp (K 1 ) supp (K 0 ). Proof. Th argumnts for th rst v parts follow immdiatly from Dnition.1 and th orrsponding parts of Proposition 1.3. Th last part follows from Part (ii). Thus, if K 1 is an -frqunt st of itms and K 0 K 1, thn K 0 is also -frqunt. Corollary.3 For vry sts of itms K; H suh that K H l (K) w hav supp (K) =supp (H). Dnition. Lt b apolarity on th st of transations T and th st of itms I and lt L I. Th family of (`; )-xtndd losurs of L rlativ to is th olltion of sts (L) =fl (L) [ H jjhj = `; H \ l (L) =;; and supp(l [ H) g: Not that ; if supp(l) CL 0; <; (L) = fl (L)g; othrwis. CL`; Furthr, CL 1; (L) =fl (L) [fig ji 6 l (L) and supp(l [fig) g: W rfr to any mmbr of CL`; (L) asan(`; )-xtndd losur. Unlss statd othrwis w always work with (1;)-xtndd losurs; so, w will just rfr to thm as xtndd losurs. W will oftn rfr to a st of itms using th trm itmst. Dnition.5 An itmst I annot b xtndd by losur if it is l-losd and its family of xtndd losurs is ;. To omput th family of (1;)-xtndd losurs for a st of itms w nd to dn th k-matrix M [k] of th Galois onntion. Evry row of this matrix orrsponds to a st of itms of ardinality k. W assum that th st of itms is I = fi 0 ;:::;i n01 g and that th sts of itms of lngth k ar arrangd in lxiographial ordr. Th olumns of th matrix M [k] orrspond to th itms of I. IfC is a st of itms suh that jcj = k and i p I, thn M [k] C;i p, th lmnt loatd in th C-row and p-th olumn is M [k] C;i p = p (C; fi p g)=jft T j t[c; i p ]=(1;:::;1)gj: Not that th matrix M 1 is symmtri. Also, th largst valu of th ntris of th C-lin of th matrix M [k] will b found in th olumns that orrspond to th mmbrs of C and possibly in othr olumns. Th losur and family of xtndd losurs of a st of itms K an b omputd starting from th tabl assoiatd to th Galois onntion, as shown in th nxt stion.

6 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions Thorm.6 Lt =(ti R; it R) b th polarity assoiatd to th rlation R T I. W hav i l (K) if and only if M [k] K;i k = jkj. Proof. Not that maxfm [k] K;j M [k] = maxfm [k] K;j j j Ig, whr j j Ig = supp (K). For any i I suh that = K;i supp (K) and i 6 K w hav to prov that i l (K). This follows from th dnition of th valu M [k] K;i whih tlls us that i appars in all transations in whih K appars so i blongs to th losur of th itmst K. Finding all -frqunt sts of itms for a spid is an important data mining problm as it rprsnts th most omputationally xpnsiv stp in nding assoiation ruls in a databas [1]. In litratur ([1], []) th valu of is alld minimum support and -frqunt sts of itms ar alld larg itmsts. In th following stions w prsnt on algorithm (Closur) for nding all - frqunt itmsts and two algorithms (MaxClosur, AltMaxClosur) for nding all maximal -frqunt itmsts. 3 Th Closur Algorithm W us th notions of losur and xtndd losur to improv on th numbr of databas sans don by th Apriori algorithm. Our algorithm is basd on th following rsult: Thorm 3.1 Lt F ;k b th olltion [ of -frqunt k-itmsts. Dn 1; CF ;k = fl (F )g[cl (F ): F F ;k Thn, for vry G F ;k+1 thr is a st C CF ;k suh that G C. Proof. Suppos G F ;k+1 ; thn, all substs of G that hav ardinality k ar inludd in F ;k.w an writ G = F S fig whr F F ;k and i is an itm. W now hav two ass: 1. If th itm i appars in all transations in whih F appars, thn i l (F) and sin l (F ) is on of th lmnts of CF ;k it follows that G must b inludd in on of th lmnts of CF ;k.. If i dos not appar in all transations in whih F appars, but it appars in a fration of transations gratr than, thn this will nsur that on (1;)- xtndd losur of F will ontain i whih mans that G must b inludd in on of th lmnts of CF ;k. Th algorithm basd on this ida uss th matrix M prsntd bfor and works as follows: Lt Maximal b th olltion of all -maximal itmsts.

7 66 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions... Candidat = all 1 itmsts KFrqunt = mpty Maximal = mpty whil (tru) do initializ th lmnts of matrix M with 0; ount th support for all lmnts of Candidat and updat thir orrsponding rows in M; for all itmsts C in Candidat do if (C is not frqunt) ontinu to nxt itmst; add C to KFrqunt; omput losur of C in l(c); omput all xtndd losurs of C and add thm to Maximal; if no xtndd losurs wr found add l(c) to Maximal; mpty Candidat; gnrat nw andidat itmsts in Candidat basd on KFrqunt; if Candidat is mpty brak; mpty KFrqunt; for all itmsts L in Maximal mark as frqunt all itmsts from Candidat that ar ontaind in L; for all itmsts C in Candidat if C is markd as frqunt add C to KFrqunt; mpty Candidat; gnrat nw andidat itmsts in Candidat basd on KFrqunt; if Candidat is mpty brak; mpty KFrqunt; rturn Maximal Not that th Closur algorithm rquirs (b n + 1) sans of th databas whr n is th dimnsion of th longst frqunt itmst. Th MaxClosur Algorithm W wrot th MaxClosur algorithm in ordr to provid a fast algorithm for nding th maximal frqunt sts. As mntiond in [3], nding all frqunt sts is an xtrmly xpnsiv omputation for som databass, so it maks sns to hav instad a fast algorithm that would allow on to know what th maximal itmsts ar (whih also mans what th frqunt itmsts ar) without nding th support for all frqunt andidats. Basd on this knowldg, on ould guid th mining pross to xtrat only th ruls on is intrstd in.

8 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions Th MaxClosur algorithm is a spialization of th Closur algorithm. It uss a similar matrix strutur for omputing th losurs and xtndd losurs and it maks us of th fat that if at any point a frqunt itmst has an mpty xtndd losur, thn that itmst is a maximal frqunt itmst (it blongs to th positiv bordr [6]). Th matrix usd in this algorithm dirs from th on dsribd bfor in th fat that its rows an orrspond to itmsts of dirnt ardinality. Thorm.1 If a frqunt itmst F annot b xtndd by losur thn it is a maximal frqunt itmst. Proof. Suppos that F annot b xtndd by losur but is not a maximal frqunt itmst. Thn, thr xists a frqunt itmst G suh that F G and G 0 F 6= ;. Lt i b an itm suh that i G and i 6 F. Sin G is frqunt it folows that i is frqunt and appars with F in a fration of transations gratr than. Thn i should ithr blong to th losur of F or to an xtndd losur of F and w hav a ontradition sin F ould not b xtndd by losur. Th algorithm's psudo is prsntd blow. It uss thr olltions: Candidat to kp th andidat maximal frqunt itmsts, Frqunt whih is usd for intrmdiat storag, and Maximal whih ontains th maximal frqunt itmsts found at any point. Candidat = all 1-itmsts; Frqunt = mpty; Maximal = mpty; whil (tru) do initializ th lmnts of matrix M with 0; ount th support for all lmnts of Candidat and updat thir orrsponding rows in M; for all itmsts C in Candidat do if C is not frqunt ontinu to nxt itmst; omput losur of C in l(c); for all xtndd losurs xl(c) of C if xl(c) has not bn alrady addd to Frqunt add xl(c) to Frqunt; if no xtndd losurs wr found if l(c) has not bn alrady addd to Maximal add l(c) to Maximal; if Frqunt is mpty brak; mpty Candidat; mov into Candidat all lmnts from Frqunt; rturn Maximal; Not that th MaxClosur algorithm rquirs in th worst as n 0 1 sans of th databas whr n is th dimnsion of th longst frqunt itmst. Th

9 68 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions... rason is that in th worst as, a andidat itmst an b xtndd with only on itm during on san of th databas so it will tak n 0 1 sans of th databas to gnrat an n-itmst. 5 An Extnsion of th MaxClosur Algorithm Th algorithm prsntd in this stion is dsignd to guarant fwr passs ovr th databas, as ompard to MaxClosur. AltMaxClosur rquirs (b n +1) sans of th databas whr n is th dimnsion of th longst frqunt itmst, th sam numbr that Closur dos. AltMaxClosur is an xtnsion of MaxClosur. Th ida is that instad of starting with a andidat C, omputing th xtndd losur of C and thn using this xtndd losur as a andidat, w an try to xtnd vn furthr th xtndd losur and us this xtnsion as a andidat. This nsurs that w will do half th numbr of passs rquird by MaxClosur. Th psudo for th algorithm is prsntd blow. It uss on additional olltion: XPrvCandidat to kp th xtndd losurs gnratd from th prvious andidat itmsts. Candidat = all 1-itmsts; Frqunt = mpty; XPrvCandidat = mpty; Maximal = mpty; whil (tru) do initializ th lmnts of matrix M with 0; ount th support for all lmnts of Candidat and updat thir orrsponding rows in M; if XPrvCandidat is not mpty do for all itmsts C in Candidat mark as not bing maximal all itmsts from XPrvCandidat that ar ontaind in C; for all itmsts L in Maximal mark as not bing maximal all itmsts from XPrvCandidat that ar ontaind in L; for all itmsts P in XPrvCandidat if P is markd as bing maximal add P to Maximal; mpty XPrvCandidat; for all itmsts C in Candidat do if C is not frqunt ontinu to nxt itmst; omput losur of C in l(c); for all xtndd losurs xl(c) of C if xl(c) has not bn alrady addd to XPrvCandidat do add xl(c) to XPrvCandidat;

10 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions for all itms i not in xl(c) suh that M(C,i) >= psilon do rat itmst C' from th union of xl(c) and i; if C' has not bn alrady addd to Frqunt add C' to Frqunt; if no xtndd losurs wr found if l(c) has not bn alrady addd to Maximal add l(c) to Maximal; if Frqunt is mpty brak; mpty Candidat; mov into Candidat all lmnts from Frqunt; rturn Maximal; Givn an xtndd losur of a andidat C, w try to xtnd it by looking at all itms that appar with C in a numbr of transations gratr or qual to. W also stor xl(c) in XPrvCandidat in th vntuality that non of ths xtnsions prov to b frqunt whih would man that xl(c) an not b furthr xtndd so it is maximal. A nw stp has bn intrud for vrifying whthr any of th lmnts of XPrvCandidat ar maximal. 6 Exprimntal rsults W implmntd Apriori, Closur, MaxClosur, and AltMaxClosur in C++. W hav usd th hash-tr strutur intrud in [] in ordr to intly tst for itmst inlusions. Th data was gnratd using th IBM Qust synthti data gnrator and was stord in binary ls. As obsrvd by othr rsarhrs [7], whn working with ls th bottlnk is in th CPU omputations, not in assing th disk. For this rason, th dirn btwn Closur and Apriori is not as big as on would xpt. In a prvious xprimnt in whih w ran a prliminary vrsion of th Apriori and Closur algorithms using Oral databass, th rsults showd a 50% improvmnt in tim on bhalf of th Closur algorithm. Our rst xprimnts usd databass with an avrag numbr of itms pr transation qual to 10, and a total of 100 itms. W hav gnratd a 50 thousand, a 100 thousand and a 1 million rows databas in ordr to hk for th salability of ah algorithm. Th rsults of th xprimnts, prsntd in Figur 1, show larly that all algorithms sal linarly with th siz of th databas. Th graphs in Figur show that at rasonabl lvls of support, our algorithms outprform th Apriori algorithm. Similar rsults hold for AltMaxClosur whn ompard to Apriori, as shown in Figur 3. On th othr hand, undr our urrnt implmntations, whih ar basd on raw ls rathr than on rlational databass, AltMaxClosur is always slowr than MaxClosur, dspit th fat that it dos fwr passs ovr th data. Th

11 70 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions... Support % Tim in sonds Apriori Closur MaxClosur AltMaxClosur 50 thousand rords thousand rords million rords Figur 1: Prforman of Algorithms thousand rows Apriori Closur MaxClosur Tim (s) Minimum support (%) Figur : Prforman Graphs for Apriori, Closur, and MaxClosur

12 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions Tim00 (s) thousand rows Apriori MaxClosur Minimum support (%) 3 Figur 3: Prforman Graphs for Apriori and AltMaxClosur rason is that AltMaxClosur has to do svral quality and inlusion tsts in ordr to nsur that no dupliats ar addd to its various olltions and thos oprations prov to b quit xpnsiv. In Figur w prsnt th rsults obtaind whn w ran th algorithms on a 100,000 thousand rows databas having an avrag of 5 itms pr transation. In this xprimnt Closur prforms slightly wors than Apriori baus it dos mor omputations whih ar not ompnsatd by th drasd numbr of passs ovr th databas. MaxClosur ontinus to prform bttr than th othr two although th improvmnt oms from ounting th support for fwr itmsts rathr than from doing fwr sans ovr th databas. For 5% minimum support th algorithms disovr 1699 maximal frqunt itmsts. W also did a tst on a databas with 100,000 transations and an avrag of 50 itms pr transation. W ran th algorithms for 50% minimum support and w obtaind 3 maximal frqunt itmsts: Apriori Closur MaxClosur AltMaxClosur Tim Databas passs 8 7 W onlud that as th avrag numbr of itms pr transation inrass, th prforman of th Closur algorithm drass. MaxClosur prforms its worst as numbr of passs but it oms prtty los to Closur baus it dos not omput all frqunt itmsts.

13 7 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions Tim (s) thousand rows Apriori Closur MaxClosur Minimum support (%) Tim (s) thousand rows Apriori MaxClosur Minimum support (%) Figur : Prforman Graphs at 5 Itms pr Transation 7 Conlusions W intrud thr nw algorithms: Closur, MaxClosur and AltMaxClosur that mak us of Galois onntions. Th xprimnts w hav don show that Closur and MaxClosur prform fastr than th standard Apriori algorithm whn running on databass with an avrag numbr of itms pr transation of lss than 5% of th total numbr of itms. For highr avrags (lik 50%), our algorithms ar outprformd by Apriori. Howvr, if databas ass is xpnsiv, Closur might still prform bttr than Apriori du to th fwr numbr of passs ovr th databas it rquirs. AltMaxClosur may b an altrnativ

14 Cristofor D., Cristofor L., Simovii D.A.: Galois Conntions for MaxClosur whn databas ass is xpnsiv; howvr, its prforman drass vry fast as th minimum support taks smallr valus. Rfrns [1] Agrawal R., Imilinski T., Swami A.:Mining Assoiation ruls btwn Sts of Itms in Larg Databass, SIGMOD 1993, pags [] Agrawal R., Srikant R.:Fast Algorithms for mining assoiation ruls, VLDB 199. [3] Bayardo R. J.:Eintly Mining Long Pattrns from Databass, SIGMOD 1998, pags [] Birkho G.:Latti Thory, Amrian Mathmatial Soity, Colloquium Publiations, 3rd dition, [5] Girz G.t al.:a Compndium of Continuous Lattis, Springr-Vrlag, Brlin, [6] Gunopulos D., Khardon R., Mannila H., Toivonn H.:Data mining, Hyprgraph Transvrsals, and Mahin Larning, PODS 1997, pags [7] Mullr A.:Fast Squntial and Paralll Algorithms for Assoiation Rul Mining: A Comparison, CS-TR-3515, Univrsity of Maryland-Collg Park, 1995.

Homework #3. 1 x. dx. It therefore follows that a sum of the

Homework #3. 1 x. dx. It therefore follows that a sum of the Danil Cannon CS 62 / Luan March 5, 2009 Homwork # 1. Th natural logarithm is dfind by ln n = n 1 dx. It thrfor follows that a sum of th 1 x sam addnd ovr th sam intrval should b both asymptotically uppr-

More information

cycle that does not cross any edges (including its own), then it has at least

cycle that does not cross any edges (including its own), then it has at least W prov th following thorm: Thorm If a K n is drawn in th plan in such a way that it has a hamiltonian cycl that dos not cross any dgs (including its own, thn it has at last n ( 4 48 π + O(n crossings Th

More information

Construction of asymmetric orthogonal arrays of strength three via a replacement method

Construction of asymmetric orthogonal arrays of strength three via a replacement method isid/ms/26/2 Fbruary, 26 http://www.isid.ac.in/ statmath/indx.php?modul=prprint Construction of asymmtric orthogonal arrays of strngth thr via a rplacmnt mthod Tian-fang Zhang, Qiaoling Dng and Alok Dy

More information

First derivative analysis

First derivative analysis Robrto s Nots on Dirntial Calculus Chaptr 8: Graphical analysis Sction First drivativ analysis What you nd to know alrady: How to us drivativs to idntiy th critical valus o a unction and its trm points

More information

CS 361 Meeting 12 10/3/18

CS 361 Meeting 12 10/3/18 CS 36 Mting 2 /3/8 Announcmnts. Homwork 4 is du Friday. If Friday is Mountain Day, homwork should b turnd in at my offic or th dpartmnt offic bfor 4. 2. Homwork 5 will b availabl ovr th wknd. 3. Our midtrm

More information

The Matrix Exponential

The Matrix Exponential Th Matrix Exponntial (with xrciss) by D. Klain Vrsion 207.0.05 Corrctions and commnts ar wlcom. Th Matrix Exponntial For ach n n complx matrix A, dfin th xponntial of A to b th matrix A A k I + A + k!

More information

Higher order derivatives

Higher order derivatives Robrto s Nots on Diffrntial Calculus Chaptr 4: Basic diffrntiation ruls Sction 7 Highr ordr drivativs What you nd to know alrady: Basic diffrntiation ruls. What you can larn hr: How to rpat th procss of

More information

Propositional Logic. Combinatorial Problem Solving (CPS) Albert Oliveras Enric Rodríguez-Carbonell. May 17, 2018

Propositional Logic. Combinatorial Problem Solving (CPS) Albert Oliveras Enric Rodríguez-Carbonell. May 17, 2018 Propositional Logic Combinatorial Problm Solving (CPS) Albrt Olivras Enric Rodríguz-Carbonll May 17, 2018 Ovrviw of th sssion Dfinition of Propositional Logic Gnral Concpts in Logic Rduction to SAT CNFs

More information

BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES

BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES DONALD M. DAVIS Abstract. If p is a prim (implicit in notation and n a positiv intgr, lt ν(n dnot th xponnt of p in n, and U(n n/p ν(n, th unit

More information

The Matrix Exponential

The Matrix Exponential Th Matrix Exponntial (with xrciss) by Dan Klain Vrsion 28928 Corrctions and commnts ar wlcom Th Matrix Exponntial For ach n n complx matrix A, dfin th xponntial of A to b th matrix () A A k I + A + k!

More information

Search sequence databases 3 10/25/2016

Search sequence databases 3 10/25/2016 Sarch squnc databass 3 10/25/2016 Etrm valu distribution Ø Suppos X is a random variabl with probability dnsity function p(, w sampl a larg numbr S of indpndnt valus of X from this distribution for an

More information

COHORT MBA. Exponential function. MATH review (part2) by Lucian Mitroiu. The LOG and EXP functions. Properties: e e. lim.

COHORT MBA. Exponential function. MATH review (part2) by Lucian Mitroiu. The LOG and EXP functions. Properties: e e. lim. MTH rviw part b Lucian Mitroiu Th LOG and EXP functions Th ponntial function p : R, dfind as Proprtis: lim > lim p Eponntial function Y 8 6 - -8-6 - - X Th natural logarithm function ln in US- log: function

More information

COUNTING TAMELY RAMIFIED EXTENSIONS OF LOCAL FIELDS UP TO ISOMORPHISM

COUNTING TAMELY RAMIFIED EXTENSIONS OF LOCAL FIELDS UP TO ISOMORPHISM COUNTING TAMELY RAMIFIED EXTENSIONS OF LOCAL FIELDS UP TO ISOMORPHISM Jim Brown Dpartmnt of Mathmatical Scincs, Clmson Univrsity, Clmson, SC 9634, USA jimlb@g.clmson.du Robrt Cass Dpartmnt of Mathmatics,

More information

BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES. 1. Statement of results

BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES. 1. Statement of results BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES DONALD M. DAVIS Abstract. If p is a prim and n a positiv intgr, lt ν p (n dnot th xponnt of p in n, and u p (n n/p νp(n th unit part of n. If α

More information

Linked-List Implementation. Linked-lists for two sets. Multiple Operations. UNION Implementation. An Application of Disjoint-Set 1/9/2014

Linked-List Implementation. Linked-lists for two sets. Multiple Operations. UNION Implementation. An Application of Disjoint-Set 1/9/2014 Disjoint Sts Data Strutur (Chap. 21) A disjoint-st is a olltion ={S 1, S 2,, S k } o distint dynami sts. Eah st is idntiid by a mmbr o th st, alld rprsntativ. Disjoint st oprations: MAKE-SET(x): rat a

More information

AP Calculus BC Problem Drill 16: Indeterminate Forms, L Hopital s Rule, & Improper Intergals

AP Calculus BC Problem Drill 16: Indeterminate Forms, L Hopital s Rule, & Improper Intergals AP Calulus BC Problm Drill 6: Indtrminat Forms, L Hopital s Rul, & Impropr Intrgals Qustion No. of Instrutions: () Rad th problm and answr hois arfully () Work th problms on papr as ndd () Pik th answr

More information

Utilizing exact and Monte Carlo methods to investigate properties of the Blume Capel Model applied to a nine site lattice.

Utilizing exact and Monte Carlo methods to investigate properties of the Blume Capel Model applied to a nine site lattice. Utilizing xat and Mont Carlo mthods to invstigat proprtis of th Blum Capl Modl applid to a nin sit latti Nik Franios Writing various xat and Mont Carlo omputr algorithms in C languag, I usd th Blum Capl

More information

u x v x dx u x v x v x u x dx d u x v x u x v x dx u x v x dx Integration by Parts Formula

u x v x dx u x v x v x u x dx d u x v x u x v x dx u x v x dx Integration by Parts Formula 7. Intgration by Parts Each drivativ formula givs ris to a corrsponding intgral formula, as w v sn many tims. Th drivativ product rul yilds a vry usful intgration tchniqu calld intgration by parts. Starting

More information

CPSC 665 : An Algorithmist s Toolkit Lecture 4 : 21 Jan Linear Programming

CPSC 665 : An Algorithmist s Toolkit Lecture 4 : 21 Jan Linear Programming CPSC 665 : An Algorithmist s Toolkit Lctur 4 : 21 Jan 2015 Lcturr: Sushant Sachdva Linar Programming Scrib: Rasmus Kyng 1. Introduction An optimization problm rquirs us to find th minimum or maximum) of

More information

Week 3: Connected Subgraphs

Week 3: Connected Subgraphs Wk 3: Connctd Subgraphs Sptmbr 19, 2016 1 Connctd Graphs Path, Distanc: A path from a vrtx x to a vrtx y in a graph G is rfrrd to an xy-path. Lt X, Y V (G). An (X, Y )-path is an xy-path with x X and y

More information

SCHUR S THEOREM REU SUMMER 2005

SCHUR S THEOREM REU SUMMER 2005 SCHUR S THEOREM REU SUMMER 2005 1. Combinatorial aroach Prhas th first rsult in th subjct blongs to I. Schur and dats back to 1916. On of his motivation was to study th local vrsion of th famous quation

More information

Abstract Interpretation: concrete and abstract semantics

Abstract Interpretation: concrete and abstract semantics Abstract Intrprtation: concrt and abstract smantics Concrt smantics W considr a vry tiny languag that manags arithmtic oprations on intgrs valus. Th (concrt) smantics of th languags cab b dfind by th funzcion

More information

Basic Polyhedral theory

Basic Polyhedral theory Basic Polyhdral thory Th st P = { A b} is calld a polyhdron. Lmma 1. Eithr th systm A = b, b 0, 0 has a solution or thr is a vctorπ such that π A 0, πb < 0 Thr cass, if solution in top row dos not ist

More information

Note If the candidate believes that e x = 0 solves to x = 0 or gives an extra solution of x = 0, then withhold the final accuracy mark.

Note If the candidate believes that e x = 0 solves to x = 0 or gives an extra solution of x = 0, then withhold the final accuracy mark. . (a) Eithr y = or ( 0, ) (b) Whn =, y = ( 0 + ) = 0 = 0 ( + ) = 0 ( )( ) = 0 Eithr = (for possibly abov) or = A 3. Not If th candidat blivs that = 0 solvs to = 0 or givs an tra solution of = 0, thn withhold

More information

(Upside-Down o Direct Rotation) β - Numbers

(Upside-Down o Direct Rotation) β - Numbers Amrican Journal of Mathmatics and Statistics 014, 4(): 58-64 DOI: 10593/jajms0140400 (Upsid-Down o Dirct Rotation) β - Numbrs Ammar Sddiq Mahmood 1, Shukriyah Sabir Ali,* 1 Dpartmnt of Mathmatics, Collg

More information

Association (Part II)

Association (Part II) Association (Part II) nanopoulos@ismll.d Outlin Improving Apriori (FP Growth, ECLAT) Qustioning confidnc masur Qustioning support masur 2 1 FP growth Algorithm Us a comprssd rprsntation of th dtb databas

More information

The second condition says that a node α of the tree has exactly n children if the arity of its label is n.

The second condition says that a node α of the tree has exactly n children if the arity of its label is n. CS 6110 S14 Hanout 2 Proof of Conflunc 27 January 2014 In this supplmntary lctur w prov that th λ-calculus is conflunt. This is rsult is u to lonzo Church (1903 1995) an J. arkly Rossr (1907 1989) an is

More information

The van der Waals interaction 1 D. E. Soper 2 University of Oregon 20 April 2012

The van der Waals interaction 1 D. E. Soper 2 University of Oregon 20 April 2012 Th van dr Waals intraction D. E. Sopr 2 Univrsity of Orgon 20 pril 202 Th van dr Waals intraction is discussd in Chaptr 5 of J. J. Sakurai, Modrn Quantum Mchanics. Hr I tak a look at it in a littl mor

More information

On spanning trees and cycles of multicolored point sets with few intersections

On spanning trees and cycles of multicolored point sets with few intersections On spanning trs and cycls of multicolord point sts with fw intrsctions M. Kano, C. Mrino, and J. Urrutia April, 00 Abstract Lt P 1,..., P k b a collction of disjoint point sts in R in gnral position. W

More information

Function Spaces. a x 3. (Letting x = 1 =)) a(0) + b + c (1) = 0. Row reducing the matrix. b 1. e 4 3. e 9. >: (x = 1 =)) a(0) + b + c (1) = 0

Function Spaces. a x 3. (Letting x = 1 =)) a(0) + b + c (1) = 0. Row reducing the matrix. b 1. e 4 3. e 9. >: (x = 1 =)) a(0) + b + c (1) = 0 unction Spacs Prrquisit: Sction 4.7, Coordinatization n this sction, w apply th tchniqus of Chaptr 4 to vctor spacs whos lmnts ar functions. Th vctor spacs P n and P ar familiar xampls of such spacs. Othr

More information

Brief Introduction to Statistical Mechanics

Brief Introduction to Statistical Mechanics Brif Introduction to Statistical Mchanics. Purpos: Ths nots ar intndd to provid a vry quick introduction to Statistical Mchanics. Th fild is of cours far mor vast than could b containd in ths fw pags.

More information

Junction Tree Algorithm 1. David Barber

Junction Tree Algorithm 1. David Barber Juntion Tr Algorithm 1 David Barbr Univrsity Collg London 1 Ths slids aompany th book Baysian Rasoning and Mahin Larning. Th book and dmos an b downloadd from www.s.ul.a.uk/staff/d.barbr/brml. Fdbak and

More information

Introduction to Arithmetic Geometry Fall 2013 Lecture #20 11/14/2013

Introduction to Arithmetic Geometry Fall 2013 Lecture #20 11/14/2013 18.782 Introduction to Arithmtic Gomtry Fall 2013 Lctur #20 11/14/2013 20.1 Dgr thorm for morphisms of curvs Lt us rstat th thorm givn at th nd of th last lctur, which w will now prov. Thorm 20.1. Lt φ:

More information

A Propagating Wave Packet Group Velocity Dispersion

A Propagating Wave Packet Group Velocity Dispersion Lctur 8 Phys 375 A Propagating Wav Packt Group Vlocity Disprsion Ovrviw and Motivation: In th last lctur w lookd at a localizd solution t) to th 1D fr-particl Schrödingr quation (SE) that corrsponds to

More information

International Journal of Scientific & Engineering Research, Volume 6, Issue 7, July ISSN

International Journal of Scientific & Engineering Research, Volume 6, Issue 7, July ISSN Intrnational Journal of Scintific & Enginring Rsarch, Volum 6, Issu 7, July-25 64 ISSN 2229-558 HARATERISTIS OF EDGE UTSET MATRIX OF PETERSON GRAPH WITH ALGEBRAI GRAPH THEORY Dr. G. Nirmala M. Murugan

More information

Lecture 14 (Oct. 30, 2017)

Lecture 14 (Oct. 30, 2017) Ltur 14 8.31 Quantum Thory I, Fall 017 69 Ltur 14 (Ot. 30, 017) 14.1 Magnti Monopols Last tim, w onsidrd a magnti fild with a magnti monopol onfiguration, and bgan to approah dsribing th quantum mhanis

More information

Final Exam Solutions

Final Exam Solutions CS 2 Advancd Data Structurs and Algorithms Final Exam Solutions Jonathan Turnr /8/20. (0 points) Suppos that r is a root of som tr in a Fionacci hap. Assum that just for a dltmin opration, r has no childrn

More information

Some Results on Interval Valued Fuzzy Neutrosophic Soft Sets ISSN

Some Results on Interval Valued Fuzzy Neutrosophic Soft Sets ISSN Som Rsults on ntrval Valud uzzy Nutrosophi Soft Sts SSN 239-9725. rokiarani Dpartmnt of Mathmatis Nirmala ollg for Womn oimbator amilnadu ndia. R. Sumathi Dpartmnt of Mathmatis Nirmala ollg for Womn oimbator

More information

1 Minimum Cut Problem

1 Minimum Cut Problem CS 6 Lctur 6 Min Cut and argr s Algorithm Scribs: Png Hui How (05), Virginia Dat: May 4, 06 Minimum Cut Problm Today, w introduc th minimum cut problm. This problm has many motivations, on of which coms

More information

CS553 Lecture Register Allocation I 3

CS553 Lecture Register Allocation I 3 Low-Lvl Issus Last ltur Intrproural analysis Toay Start low-lvl issus Rgistr alloation Latr Mor rgistr alloation Instrution shuling CS553 Ltur Rgistr Alloation I 2 Rgistr Alloation Prolm Assign an unoun

More information

Mor Tutorial at www.dumblittldoctor.com Work th problms without a calculator, but us a calculator to chck rsults. And try diffrntiating your answrs in part III as a usful chck. I. Applications of Intgration

More information

Steinberg s Conjecture is false

Steinberg s Conjecture is false Stinrg s Conjtur is als arxiv:1604.05108v2 [math.co] 19 Apr 2016 Vinnt Cohn-Aa Mihal Hig Danil Král Zhntao Li Estan Salgao Astrat Stinrg onjtur in 1976 that vry planar graph with no yls o lngth our or

More information

Derangements and Applications

Derangements and Applications 2 3 47 6 23 Journal of Intgr Squncs, Vol. 6 (2003), Articl 03..2 Drangmnts and Applications Mhdi Hassani Dpartmnt of Mathmatics Institut for Advancd Studis in Basic Scincs Zanjan, Iran mhassani@iasbs.ac.ir

More information

ON RIGHT(LEFT) DUO PO-SEMIGROUPS. S. K. Lee and K. Y. Park

ON RIGHT(LEFT) DUO PO-SEMIGROUPS. S. K. Lee and K. Y. Park Kangwon-Kyungki Math. Jour. 11 (2003), No. 2, pp. 147 153 ON RIGHT(LEFT) DUO PO-SEMIGROUPS S. K. L and K. Y. Park Abstract. W invstigat som proprtis on right(rsp. lft) duo po-smigroups. 1. Introduction

More information

Formal Concept Analysis

Formal Concept Analysis Forml Conpt Anlysis Conpt intnts s losd sts Closur Systms nd Implitions 4 Closur Systms 0.06.005 Nxt-Closur ws dvlopd y B. Gntr (984). Lt M = {,..., n}. A M is ltilly smllr thn B M, if B A if th smllst

More information

Section 11.6: Directional Derivatives and the Gradient Vector

Section 11.6: Directional Derivatives and the Gradient Vector Sction.6: Dirctional Drivativs and th Gradint Vctor Practic HW rom Stwart Ttbook not to hand in p. 778 # -4 p. 799 # 4-5 7 9 9 35 37 odd Th Dirctional Drivativ Rcall that a b Slop o th tangnt lin to th

More information

COMPUTER GENERATED HOLOGRAMS Optical Sciences 627 W.J. Dallas (Monday, April 04, 2005, 8:35 AM) PART I: CHAPTER TWO COMB MATH.

COMPUTER GENERATED HOLOGRAMS Optical Sciences 627 W.J. Dallas (Monday, April 04, 2005, 8:35 AM) PART I: CHAPTER TWO COMB MATH. C:\Dallas\0_Courss\03A_OpSci_67\0 Cgh_Book\0_athmaticalPrliminaris\0_0 Combath.doc of 8 COPUTER GENERATED HOLOGRAS Optical Scincs 67 W.J. Dallas (onday, April 04, 005, 8:35 A) PART I: CHAPTER TWO COB ATH

More information

y = 2xe x + x 2 e x at (0, 3). solution: Since y is implicitly related to x we have to use implicit differentiation: 3 6y = 0 y = 1 2 x ln(b) ln(b)

y = 2xe x + x 2 e x at (0, 3). solution: Since y is implicitly related to x we have to use implicit differentiation: 3 6y = 0 y = 1 2 x ln(b) ln(b) 4. y = y = + 5. Find th quation of th tangnt lin for th function y = ( + ) 3 whn = 0. solution: First not that whn = 0, y = (1 + 1) 3 = 8, so th lin gos through (0, 8) and thrfor its y-intrcpt is 8. y

More information

ECE602 Exam 1 April 5, You must show ALL of your work for full credit.

ECE602 Exam 1 April 5, You must show ALL of your work for full credit. ECE62 Exam April 5, 27 Nam: Solution Scor: / This xam is closd-book. You must show ALL of your work for full crdit. Plas rad th qustions carfully. Plas chck your answrs carfully. Calculators may NOT b

More information

UNTYPED LAMBDA CALCULUS (II)

UNTYPED LAMBDA CALCULUS (II) 1 UNTYPED LAMBDA CALCULUS (II) RECALL: CALL-BY-VALUE O.S. Basic rul Sarch ruls: (\x.) v [v/x] 1 1 1 1 v v CALL-BY-VALUE EVALUATION EXAMPLE (\x. x x) (\y. y) x x [\y. y / x] = (\y. y) (\y. y) y [\y. y /

More information

perm4 A cnt 0 for for if A i 1 A i cnt cnt 1 cnt i j. j k. k l. i k. j l. i l

perm4 A cnt 0 for for if A i 1 A i cnt cnt 1 cnt i j. j k. k l. i k. j l. i l h 4D, 4th Rank, Antisytric nsor and th 4D Equivalnt to th Cross Product or Mor Fun with nsors!!! Richard R Shiffan Digital Graphics Assoc 8 Dunkirk Av LA, Ca 95 rrs@isidu his docunt dscribs th four dinsional

More information

MCB137: Physical Biology of the Cell Spring 2017 Homework 6: Ligand binding and the MWC model of allostery (Due 3/23/17)

MCB137: Physical Biology of the Cell Spring 2017 Homework 6: Ligand binding and the MWC model of allostery (Due 3/23/17) MCB37: Physical Biology of th Cll Spring 207 Homwork 6: Ligand binding and th MWC modl of allostry (Du 3/23/7) Hrnan G. Garcia March 2, 207 Simpl rprssion In class, w drivd a mathmatical modl of how simpl

More information

EEO 401 Digital Signal Processing Prof. Mark Fowler

EEO 401 Digital Signal Processing Prof. Mark Fowler EEO 401 Digital Signal Procssing Prof. Mark Fowlr Dtails of th ot St #19 Rading Assignmnt: Sct. 7.1.2, 7.1.3, & 7.2 of Proakis & Manolakis Dfinition of th So Givn signal data points x[n] for n = 0,, -1

More information

Examples and applications on SSSP and MST

Examples and applications on SSSP and MST Exampls an applications on SSSP an MST Dan (Doris) H & Junhao Gan ITEE Univrsity of Qunslan COMP3506/7505, Uni of Qunslan Exampls an applications on SSSP an MST Dijkstra s Algorithm Th algorithm solvs

More information

Section 6.1. Question: 2. Let H be a subgroup of a group G. Then H operates on G by left multiplication. Describe the orbits for this operation.

Section 6.1. Question: 2. Let H be a subgroup of a group G. Then H operates on G by left multiplication. Describe the orbits for this operation. MAT 444 H Barclo Spring 004 Homwork 6 Solutions Sction 6 Lt H b a subgroup of a group G Thn H oprats on G by lft multiplication Dscrib th orbits for this opration Th orbits of G ar th right costs of H

More information

Where k is either given or determined from the data and c is an arbitrary constant.

Where k is either given or determined from the data and c is an arbitrary constant. Exponntial growth and dcay applications W wish to solv an quation that has a drivativ. dy ky k > dx This quation says that th rat of chang of th function is proportional to th function. Th solution is

More information

Searching Linked Lists. Perfect Skip List. Building a Skip List. Skip List Analysis (1) Assume the list is sorted, but is stored in a linked list.

Searching Linked Lists. Perfect Skip List. Building a Skip List. Skip List Analysis (1) Assume the list is sorted, but is stored in a linked list. 3 3 4 8 6 3 3 4 8 6 3 3 4 8 6 () (d) 3 Sarching Linkd Lists Sarching Linkd Lists Sarching Linkd Lists ssum th list is sortd, but is stord in a linkd list. an w us binary sarch? omparisons? Work? What if

More information

Addition of angular momentum

Addition of angular momentum Addition of angular momntum April, 0 Oftn w nd to combin diffrnt sourcs of angular momntum to charactriz th total angular momntum of a systm, or to divid th total angular momntum into parts to valuat th

More information

Computing and Communications -- Network Coding

Computing and Communications -- Network Coding 89 90 98 00 Computing and Communications -- Ntwork Coding Dr. Zhiyong Chn Institut of Wirlss Communications Tchnology Shanghai Jiao Tong Univrsity China Lctur 5- Nov. 05 0 Classical Information Thory Sourc

More information

MA1506 Tutorial 2 Solutions. Question 1. (1a) 1 ) y x. e x. 1 exp (in general, Integrating factor is. ye dx. So ) (1b) e e. e c.

MA1506 Tutorial 2 Solutions. Question 1. (1a) 1 ) y x. e x. 1 exp (in general, Integrating factor is. ye dx. So ) (1b) e e. e c. MA56 utorial Solutions Qustion a Intgrating fator is ln p p in gnral, multipl b p So b ln p p sin his kin is all a Brnoulli quation -- st Sin w fin Y, Y Y, Y Y p Qustion Dfin v / hn our quation is v μ

More information

Differential Equations

Differential Equations Prfac Hr ar m onlin nots for m diffrntial quations cours that I tach hr at Lamar Univrsit. Dspit th fact that ths ar m class nots, th should b accssibl to anon wanting to larn how to solv diffrntial quations

More information

Quasi-Classical States of the Simple Harmonic Oscillator

Quasi-Classical States of the Simple Harmonic Oscillator Quasi-Classical Stats of th Simpl Harmonic Oscillator (Draft Vrsion) Introduction: Why Look for Eignstats of th Annihilation Oprator? Excpt for th ground stat, th corrspondnc btwn th quantum nrgy ignstats

More information

Coupled Pendulums. Two normal modes.

Coupled Pendulums. Two normal modes. Tim Dpndnt Two Stat Problm Coupld Pndulums Wak spring Two normal mods. No friction. No air rsistanc. Prfct Spring Start Swinging Som tim latr - swings with full amplitud. stationary M +n L M +m Elctron

More information

On the irreducibility of some polynomials in two variables

On the irreducibility of some polynomials in two variables ACTA ARITHMETICA LXXXII.3 (1997) On th irrducibility of som polynomials in two variabls by B. Brindza and Á. Pintér (Dbrcn) To th mmory of Paul Erdős Lt f(x) and g(y ) b polynomials with intgral cofficints

More information

4037 ADDITIONAL MATHEMATICS

4037 ADDITIONAL MATHEMATICS CAMBRIDGE INTERNATIONAL EXAMINATIONS GCE Ordinary Lvl MARK SCHEME for th Octobr/Novmbr 0 sris 40 ADDITIONAL MATHEMATICS 40/ Papr, maimum raw mark 80 This mark schm is publishd as an aid to tachrs and candidats,

More information

Physics 506 Winter 2006 Homework Assignment #12 Solutions. Textbook problems: Ch. 14: 14.2, 14.4, 14.6, 14.12

Physics 506 Winter 2006 Homework Assignment #12 Solutions. Textbook problems: Ch. 14: 14.2, 14.4, 14.6, 14.12 Physis 56 Wintr 6 Homwork Assignmnt # Solutions Ttbook problms: Ch. 4: 4., 4.4, 4.6, 4. 4. A partil of harg is moving in narly uniform nonrlativisti motion. For tims nar t = t, its vtorial position an

More information

A. Limits and Horizontal Asymptotes ( ) f x f x. f x. x "±# ( ).

A. Limits and Horizontal Asymptotes ( ) f x f x. f x. x ±# ( ). A. Limits and Horizontal Asymptots What you ar finding: You can b askd to find lim x "a H.A.) problm is asking you find lim x "# and lim x "$#. or lim x "±#. Typically, a horizontal asymptot algbraically,

More information

22/ Breakdown of the Born-Oppenheimer approximation. Selection rules for rotational-vibrational transitions. P, R branches.

22/ Breakdown of the Born-Oppenheimer approximation. Selection rules for rotational-vibrational transitions. P, R branches. Subjct Chmistry Papr No and Titl Modul No and Titl Modul Tag 8/ Physical Spctroscopy / Brakdown of th Born-Oppnhimr approximation. Slction ruls for rotational-vibrational transitions. P, R branchs. CHE_P8_M

More information

Sundials and Linear Algebra

Sundials and Linear Algebra Sundials and Linar Algbra M. Scot Swan July 2, 25 Most txts on crating sundials ar dirctd towards thos who ar solly intrstd in making and using sundials and usually assums minimal mathmatical background.

More information

Addition of angular momentum

Addition of angular momentum Addition of angular momntum April, 07 Oftn w nd to combin diffrnt sourcs of angular momntum to charactriz th total angular momntum of a systm, or to divid th total angular momntum into parts to valuat

More information

Abstract Interpretation. Lecture 5. Profs. Aiken, Barrett & Dill CS 357 Lecture 5 1

Abstract Interpretation. Lecture 5. Profs. Aiken, Barrett & Dill CS 357 Lecture 5 1 Abstract Intrprtation 1 History On brakthrough papr Cousot & Cousot 77 (?) Inspird by Dataflow analysis Dnotational smantics Enthusiastically mbracd by th community At last th functional community... At

More information

Self-Adjointness and Its Relationship to Quantum Mechanics. Ronald I. Frank 2016

Self-Adjointness and Its Relationship to Quantum Mechanics. Ronald I. Frank 2016 Ronald I. Frank 06 Adjoint https://n.wikipdia.org/wiki/adjoint In gnral thr is an oprator and a procss that dfin its adjoint *. It is thn slf-adjoint if *. Innr product spac https://n.wikipdia.org/wiki/innr_product_spac

More information

EXST Regression Techniques Page 1

EXST Regression Techniques Page 1 EXST704 - Rgrssion Tchniqus Pag 1 Masurmnt rrors in X W hav assumd that all variation is in Y. Masurmnt rror in this variabl will not ffct th rsults, as long as thy ar uncorrlatd and unbiasd, sinc thy

More information

10. Limits involving infinity

10. Limits involving infinity . Limits involving infinity It is known from th it ruls for fundamntal arithmtic oprations (+,-,, ) that if two functions hav finit its at a (finit or infinit) point, that is, thy ar convrgnt, th it of

More information

u r du = ur+1 r + 1 du = ln u + C u sin u du = cos u + C cos u du = sin u + C sec u tan u du = sec u + C e u du = e u + C

u r du = ur+1 r + 1 du = ln u + C u sin u du = cos u + C cos u du = sin u + C sec u tan u du = sec u + C e u du = e u + C Tchniqus of Intgration c Donald Kridr and Dwight Lahr In this sction w ar going to introduc th first approachs to valuating an indfinit intgral whos intgrand dos not hav an immdiat antidrivativ. W bgin

More information

1973 AP Calculus AB: Section I

1973 AP Calculus AB: Section I 97 AP Calculus AB: Sction I 9 Minuts No Calculator Not: In this amination, ln dnots th natural logarithm of (that is, logarithm to th bas ).. ( ) d= + C 6 + C + C + C + C. If f ( ) = + + + and ( ), g=

More information

Aim To manage files and directories using Linux commands. 1. file Examines the type of the given file or directory

Aim To manage files and directories using Linux commands. 1. file Examines the type of the given file or directory m E x. N o. 3 F I L E M A N A G E M E N T Aim To manag ils and dirctoris using Linux commands. I. F i l M a n a g m n t 1. il Examins th typ o th givn il or dirctory i l i l n a m > ( o r ) < d i r c t

More information

Math 34A. Final Review

Math 34A. Final Review Math A Final Rviw 1) Us th graph of y10 to find approimat valus: a) 50 0. b) y (0.65) solution for part a) first writ an quation: 50 0. now tak th logarithm of both sids: log() log(50 0. ) pand th right

More information

Mathematics 1110H Calculus I: Limits, derivatives, and Integrals Trent University, Summer 2018 Solutions to the Actual Final Examination

Mathematics 1110H Calculus I: Limits, derivatives, and Integrals Trent University, Summer 2018 Solutions to the Actual Final Examination Mathmatics H Calculus I: Limits, rivativs, an Intgrals Trnt Univrsity, Summr 8 Solutions to th Actual Final Eamination Tim-spac: 9:-: in FPHL 7. Brought to you by Stfan B lan k. Instructions: Do parts

More information

Outline. 1 Introduction. 2 Min-Cost Spanning Trees. 4 Example

Outline. 1 Introduction. 2 Min-Cost Spanning Trees. 4 Example Outlin Computr Sin 33 Computtion o Minimum-Cost Spnnin Trs Prim's Alorithm Introution Mik Joson Dprtmnt o Computr Sin Univrsity o Clry Ltur #33 3 Alorithm Gnrl Constrution Mik Joson (Univrsity o Clry)

More information

1 Isoparametric Concept

1 Isoparametric Concept UNIVERSITY OF CALIFORNIA BERKELEY Dpartmnt of Civil Enginring Spring 06 Structural Enginring, Mchanics and Matrials Profssor: S. Govindj Nots on D isoparamtric lmnts Isoparamtric Concpt Th isoparamtric

More information

CS 6353 Compiler Construction, Homework #1. 1. Write regular expressions for the following informally described languages:

CS 6353 Compiler Construction, Homework #1. 1. Write regular expressions for the following informally described languages: CS 6353 Compilr Construction, Homwork #1 1. Writ rgular xprssions for th following informally dscribd languags: a. All strings of 0 s and 1 s with th substring 01*1. Answr: (0 1)*01*1(0 1)* b. All strings

More information

SECTION where P (cos θ, sin θ) and Q(cos θ, sin θ) are polynomials in cos θ and sin θ, provided Q is never equal to zero.

SECTION where P (cos θ, sin θ) and Q(cos θ, sin θ) are polynomials in cos θ and sin θ, provided Q is never equal to zero. SETION 6. 57 6. Evaluation of Dfinit Intgrals Exampl 6.6 W hav usd dfinit intgrals to valuat contour intgrals. It may com as a surpris to larn that contour intgrals and rsidus can b usd to valuat crtain

More information

Random Process Part 1

Random Process Part 1 Random Procss Part A random procss t (, ζ is a signal or wavform in tim. t : tim ζ : outcom in th sampl spac Each tim w rapat th xprimnt, a nw wavform is gnratd. ( W will adopt t for short. Tim sampls

More information

Assignment 4 Biophys 4322/5322

Assignment 4 Biophys 4322/5322 Assignmnt 4 Biophys 4322/5322 Tylr Shndruk Fbruary 28, 202 Problm Phillips 7.3. Part a R-onsidr dimoglobin utilizing th anonial nsmbl maning rdriv Eq. 3 from Phillips Chaptr 7. For a anonial nsmbl p E

More information

Fourier Transforms and the Wave Equation. Key Mathematics: More Fourier transform theory, especially as applied to solving the wave equation.

Fourier Transforms and the Wave Equation. Key Mathematics: More Fourier transform theory, especially as applied to solving the wave equation. Lur 7 Fourir Transforms and th Wav Euation Ovrviw and Motivation: W first discuss a fw faturs of th Fourir transform (FT), and thn w solv th initial-valu problm for th wav uation using th Fourir transform

More information

EECE 301 Signals & Systems Prof. Mark Fowler

EECE 301 Signals & Systems Prof. Mark Fowler EECE 301 Signals & Systms Prof. Mark Fowlr ot St #21 D-T Signals: Rlation btwn DFT, DTFT, & CTFT 1/16 W can us th DFT to implmnt numrical FT procssing This nabls us to numrically analyz a signal to find

More information

Thus, because if either [G : H] or [H : K] is infinite, then [G : K] is infinite, then [G : K] = [G : H][H : K] for all infinite cases.

Thus, because if either [G : H] or [H : K] is infinite, then [G : K] is infinite, then [G : K] = [G : H][H : K] for all infinite cases. Homwork 5 M 373K Solutions Mark Lindbrg and Travis Schdlr 1. Prov that th ring Z/mZ (for m 0) is a fild if and only if m is prim. ( ) Proof by Contrapositiv: Hr, thr ar thr cass for m not prim. m 0: Whn

More information

1997 AP Calculus AB: Section I, Part A

1997 AP Calculus AB: Section I, Part A 997 AP Calculus AB: Sction I, Part A 50 Minuts No Calculator Not: Unlss othrwis spcifid, th domain of a function f is assumd to b th st of all ral numbrs for which f () is a ral numbr.. (4 6 ) d= 4 6 6

More information

Title: Vibrational structure of electronic transition

Title: Vibrational structure of electronic transition Titl: Vibrational structur of lctronic transition Pag- Th band spctrum sn in th Ultra-Violt (UV) and visibl (VIS) rgions of th lctromagntic spctrum can not intrprtd as vibrational and rotational spctrum

More information

EEO 401 Digital Signal Processing Prof. Mark Fowler

EEO 401 Digital Signal Processing Prof. Mark Fowler EEO 401 Digital Signal Procssing Prof. Mark Fowlr ot St #18 Introduction to DFT (via th DTFT) Rading Assignmnt: Sct. 7.1 of Proakis & Manolakis 1/24 Discrt Fourir Transform (DFT) W v sn that th DTFT is

More information

Calculus II (MAC )

Calculus II (MAC ) Calculus II (MAC232-2) Tst 2 (25/6/25) Nam (PRINT): Plas show your work. An answr with no work rcivs no crdit. You may us th back of a pag if you nd mor spac for a problm. You may not us any calculators.

More information

Chapter 6 Folding. Folding

Chapter 6 Folding. Folding Chaptr 6 Folding Wintr 1 Mokhtar Abolaz Folding Th folding transformation is usd to systmatically dtrmin th control circuits in DSP architctur whr multipl algorithm oprations ar tim-multiplxd to a singl

More information

MCE503: Modeling and Simulation of Mechatronic Systems Discussion on Bond Graph Sign Conventions for Electrical Systems

MCE503: Modeling and Simulation of Mechatronic Systems Discussion on Bond Graph Sign Conventions for Electrical Systems MCE503: Modling and Simulation o Mchatronic Systms Discussion on Bond Graph Sign Convntions or Elctrical Systms Hanz ichtr, PhD Clvland Stat Univrsity, Dpt o Mchanical Enginring 1 Basic Assumption In a

More information

Linear Non-Gaussian Structural Equation Models

Linear Non-Gaussian Structural Equation Models IMPS 8, Durham, NH Linar Non-Gaussian Structural Equation Modls Shohi Shimizu, Patrik Hoyr and Aapo Hyvarinn Osaka Univrsity, Japan Univrsity of Hlsinki, Finland Abstract Linar Structural Equation Modling

More information

Difference -Analytical Method of The One-Dimensional Convection-Diffusion Equation

Difference -Analytical Method of The One-Dimensional Convection-Diffusion Equation Diffrnc -Analytical Mthod of Th On-Dimnsional Convction-Diffusion Equation Dalabav Umurdin Dpartmnt mathmatic modlling, Univrsity of orld Economy and Diplomacy, Uzbistan Abstract. An analytical diffrncing

More information

The Equitable Dominating Graph

The Equitable Dominating Graph Intrnational Journal of Enginring Rsarch and Tchnology. ISSN 0974-3154 Volum 8, Numbr 1 (015), pp. 35-4 Intrnational Rsarch Publication Hous http://www.irphous.com Th Equitabl Dominating Graph P.N. Vinay

More information

That is, we start with a general matrix: And end with a simpler matrix:

That is, we start with a general matrix: And end with a simpler matrix: DIAGON ALIZATION OF THE STR ESS TEN SOR INTRO DUCTIO N By th us of Cauchy s thorm w ar abl to rduc th numbr of strss componnts in th strss tnsor to only nin valus. An additional simplification of th strss

More information

ELECTRON-MUON SCATTERING

ELECTRON-MUON SCATTERING ELECTRON-MUON SCATTERING ABSTRACT Th lctron charg is considrd to b distributd or xtndd in spac. Th diffrntial of th lctron charg is st qual to a function of lctron charg coordinats multiplid by a four-dimnsional

More information

ph People Grade Level: basic Duration: minutes Setting: classroom or field site

ph People Grade Level: basic Duration: minutes Setting: classroom or field site ph Popl Adaptd from: Whr Ar th Frogs? in Projct WET: Curriculum & Activity Guid. Bozman: Th Watrcours and th Council for Environmntal Education, 1995. ph Grad Lvl: basic Duration: 10 15 minuts Stting:

More information