A Collaborative Filtering Recommendation Algorithm Based on Influence Sets

Σχετικά έγγραφα
2002 Journal of Software

ER-Tree (Extended R*-Tree)

ΕΡΓΑΣΙΑ ΜΑΘΗΜΑΤΟΣ: ΘΕΩΡΙΑ ΒΕΛΤΙΣΤΟΥ ΕΛΕΓΧΟΥ ΦΙΛΤΡΟ KALMAN ΜΩΥΣΗΣ ΛΑΖΑΡΟΣ

Vol. 40 No Journal of Jiangxi Normal University Natural Science Jul. 2016

J. of Math. (PRC) u(t k ) = I k (u(t k )), k = 1, 2,, (1.6) , [3, 4] (1.1), (1.2), (1.3), [6 8]

TRM +4!5"2# 6!#!-!2&'!5$27!842//22&'9&2:1*;832<

Vol. 31,No JOURNAL OF CHINA UNIVERSITY OF SCIENCE AND TECHNOLOGY Feb

Levin Lin(1992) Oh(1996),Wu(1996) Papell(1997) Im, Pesaran Shin(1996) Canzoneri, Cumby Diba(1999) Lee, Pesaran Smith(1997) FGLS SUR

Quick algorithm f or computing core attribute

Appendix. The solution begins with Eq. (2.15) from the text, which we repeat here for 1, (A.1)

Lecture 12 Modulation and Sampling

, Litrrow. Maxwell. Helmholtz Fredholm, . 40 Maystre [4 ], Goray [5 ], Kleemann [6 ] PACC: 4210, 4110H

Χρονοσειρές Μάθημα 3

Anomaly Detection with Neighborhood Preservation Principle

The Research on Sampling Estimation of Seasonal Index Based on Stratified Random Sampling

Kenta OKU and Fumio HATTORI

Study on the Strengthen Method of Masonry Structure by Steel Truss for Collapse Prevention

A research on the influence of dummy activity on float in an AOA network and its amendments

A Control Method of Errors in Long-Term Integration

Newman Modularity Newman [4], [5] Newman Q Q Q greedy algorithm[6] Newman Newman Q 1 Tabu Search[7] Newman Newman Newman Q Newman 1 2 Newman 3

Motion analysis and simulation of a stratospheric airship

Q L -BFGS. Method of Q through full waveform inversion based on L -BFGS algorithm. SUN Hui-qiu HAN Li-guo XU Yang-yang GAO Han ZHOU Yan ZHANG Pan

Positive solutions for a multi-point eigenvalue. problem involving the one dimensional

Nov Journal of Zhengzhou University Engineering Science Vol. 36 No FCM. A doi /j. issn

6.003: Signals and Systems

6.003: Signals and Systems. Modulation

( ) ( t) ( 0) ( ) dw w. = = β. Then the solution of (1.1) is easily found to. wt = t+ t. We generalize this to the following nonlinear differential

Analysis on construction application of lager diameter pile foundation engineering in Guangdong coastal areas

2002 Journal of Software

VBA Microsoft Excel. J. Comput. Chem. Jpn., Vol. 5, No. 1, pp (2006)

( ) ( ) ( ) Fourier series. ; m is an integer. r(t) is periodic (T>0), r(t+t) = r(t), t Fundamental period T 0 = smallest T. Fundamental frequency ω

Arbitrage Analysis of Futures Market with Frictions

Bundle Adjustment for 3-D Reconstruction: Implementation and Evaluation

2 ~ 8 Hz Hz. Blondet 1 Trombetti 2-4 Symans 5. = - M p. M p. s 2 x p. s 2 x t x t. + C p. sx p. + K p. x p. C p. s 2. x tp x t.

PACS: Pj, Gg

Reading Order Detection for Text Layout Excluded by Image

Approximation Expressions for the Temperature Integral

Research of Han Character Internal Codes Recognition Algorithm in the Multi2lingual Environment

Application of Object Oriented Programming to a Computational Fluid Dynamics

ΤΑΞΙΝΟΜΙΚΟΙ ΚΑΝΟΝΕΣ ΓΙΑ ΕΛΛΕΙΠΤΙΚΟΥΣ ΠΛΗΘΥΣΜΟΥΣ ΜΕ ΜΟΝΟΤΟΝΑ ΕΛΛΙΠΗ Ε ΟΜΕΝΑ

Research on real-time inverse kinematics algorithms for 6R robots

A summation formula ramified with hypergeometric function and involving recurrence relation

Μοντέλα Ανάκτησης Ι (Retrieval Models)

1 (forward modeling) 2 (data-driven modeling) e- Quest EnergyPlus DeST 1.1. {X t } ARMA. S.Sp. Pappas [4]

The Euler Equations! λ 1. λ 2. λ 3. ρ ρu. E = e + u 2 /2. E + p ρ. = de /dt. = dh / dt; h = h( T ); c p. / c v. ; γ = c p. p = ( γ 1)ρe. c v.

( ) , ) , ; kg 1) 80 % kg. Vol. 28,No. 1 Jan.,2006 RESOURCES SCIENCE : (2006) ,2 ,,,, ; ;

9.1 Introduction 9.2 Lags in the Error Term: Autocorrelation 9.3 Estimating an AR(1) Error Model 9.4 Testing for Autocorrelation 9.

Reservoir modeling. Reservoir modelling Linear reservoirs. The linear reservoir, no input. Starting up reservoir modeling

College of Life Science, Dalian Nationalities University, Dalian , PR China.


Η ΨΥΧΙΑΤΡΙΚΗ - ΨΥΧΟΛΟΓΙΚΗ ΠΡΑΓΜΑΤΟΓΝΩΜΟΣΥΝΗ ΣΤΗΝ ΠΟΙΝΙΚΗ ΔΙΚΗ

Application of Genetic Algorithm in Architectural Conceptual Design

Homework 8 Model Solution Section

LUO, Hong2Qun LIU, Shao2Pu Ξ LI, Nian2Bing

University of Washington Department of Chemistry Chemistry 553 Spring Quarter 2010 Homework Assignment 3 Due 04/26/10

Area Location and Recognition of Video Text Based on Depth Learning Method

Μηχανισμοί πρόβλεψης προσήμων σε προσημασμένα μοντέλα κοινωνικών δικτύων ΔΙΠΛΩΜΑΤΙΚΗ ΕΡΓΑΣΙΑ

Wiki. Wiki. Analysis of user activity of closed Wiki used by small groups

Quantum dot sensitized solar cells with efficiency over 12% based on tetraethyl orthosilicate additive in polysulfide electrolyte

2016 IEEE/ACM International Conference on Mobile Software Engineering and Systems

Congruence Classes of Invertible Matrices of Order 3 over F 2

Apr Vol.26 No.2. Pure and Applied Mathematics O157.5 A (2010) (d(u)d(v)) α, 1, (1969-),,.

: Monte Carlo EM 313, Louis (1982) EM, EM Newton-Raphson, /. EM, 2 Monte Carlo EM Newton-Raphson, Monte Carlo EM, Monte Carlo EM, /. 3, Monte Carlo EM

A Fault Identification Algorithm for Satellite Networks Based on System Level Diagnosis

Development of the Nursing Program for Rehabilitation of Woman Diagnosed with Breast Cancer

MA6451-PROBABILITY & RANDOM PROCESS. UNIT-IV-CORRELATION AND SPECTRAL DENSITIES By K.VIJAYALAKSHMI Dept. of Applied mathematics


ΕΚΠΑΙΔΕΥΣΗ, ΑΜΥΝΤΙΚΕΣ ΔΑΠΑΝΕΣ ΚΑΙ ΟΙΚΟΝΟΜΙΚΗ ΑΝΑΠΤΥΞΗ: ΜΙΑ ΕΜΠΕΙΡΙΚΗ ΈΡΕΥΝΑ ΓΙΑ ΤΗΝ ΚΥΠΡΟ

Correction of chromatic aberration for human eyes with diffractive-refractive hybrid elements

Anti-aliasing Prefilter (6B) Young Won Lim 6/8/12


MIDI [8] MIDI. [9] Hsu [1], [2] [10] Salamon [11] [5] Song [6] Sony, Minato, Tokyo , Japan a) b)

Resurvey of Possible Seismic Fissures in the Old-Edo River in Tokyo

Lossless Image Authentication Algorithm with Minimal Expansion

Journal of the Institute of Science and Engineering. Chuo University

Ερευνητική+Ομάδα+Τεχνολογιών+ Διαδικτύου+

Retrieval of Seismic Data Recorded on Open-reel-type Magnetic Tapes (MT) by Using Existing Devices

n 1 n 3 choice node (shelf) choice node (rough group) choice node (representative candidate)

Stress Relaxation Test and Constitutive Equation of Saturated Soft Soil

ΤΕΧΝΟΛΟΓΙΚΟ ΠΑΝΕΠΙΣΤΗΜΙΟ ΚΥΠΡΟΥ ΤΜΗΜΑ ΝΟΣΗΛΕΥΤΙΚΗΣ

Error ana lysis of P2wave non2hyperbolic m oveout veloc ity in layered media

Accounts receivable LTV ratio optimization based on supply chain credit

1530 ( ) 2014,54(12),, E (, 1, X ) [4],,, α, T α, β,, T β, c, P(T β 1 T α,α, β,c) 1 1,,X X F, X E F X E X F X F E X E 1 [1-2] , 2 : X X 1 X 2 ;

No. 7 Modular Machine Tool & Automatic Manufacturing Technique. Jul TH166 TG659 A

An Automatic Modulation Classifier using a Frequency Discriminator for Intelligent Software Defined Radio

Buried Markov Model Pairwise

Japanese Fuzzy String Matching in Cooking Recipes

Answers - Worksheet A ALGEBRA PMT. 1 a = 7 b = 11 c = 1 3. e = 0.1 f = 0.3 g = 2 h = 10 i = 3 j = d = k = 3 1. = 1 or 0.5 l =

On Strong Product of Two Fuzzy Graphs

Key Formulas From Larson/Farber Elementary Statistics: Picturing the World, Second Edition 2002 Prentice Hall

Super-Resolution Reconstruction for Face Images Based on Particle Filters Method

Shortness Ambiguity TEAM Ungrammaticality

ΤΕΧΝΟΛΟΓΙΚΟ ΠΑΝΕΠΙΣΤΗΜΙΟ ΚΥΠΡΟΥ ΣΧΟΛΗ ΕΠΙΣΤΗΜΩΝ ΥΓΕΙΑΣ ΤΜΗΜΑ ΝΟΣΗΛΕΥΤΙΚΗΣ

Ne ws content extraction based on block distribution

r t te 2t i t Find the derivative of the vector function. 19. r t e t cos t i e t sin t j ln t k Evaluate the integral.

Customized Pricing Recommender System Simple Implementation and Preliminary Experiments

Nash. Nash. Nash F224 O Hopp 5. Nash. Nash. Vol. 16 No. 3 Mar JOURNAL OF MANAGEMENT SCIENCES IN CHINA

ACTA MATHEMATICAE APPLICATAE SINICA Nov., ( µ ) ( (

ω = radians per sec, t = 3 sec

FENXI HUAXUE Chinese Journal of Analytical Chemistry. Savitzky-Golay. n = SG SG. Savitzky-Golay mmol /L 5700.

Transcript:

ISSN 1000-985, CODEN RUXUE E-mail: os@iscasaccn Journal of Sofware, Vol18, No7, July 007, 1685 1694 h://wwwosorgcn DOI: 101360/os181685 Tel/Fax: +86-10-656563 007 by Journal of Sofware All righs reserved 1+, 1 (, 510006 (, 51075 A Collaboraive Filering Recommendaion Algorihm Based on Influence Ses CHEN Jian 1+, YIN Jian 1 (School of Comuer Science and Engineering, Souh China Universiy of Technology, Guangzhou 510006, China (Dearmen of Comuer Science, Sun Ya-Se Universiy, Guangzhou 51075, China + Corresonding auhor: Phn: +86-0-33509119, Fax: +86-0-3938018, E-mail: ellachen@scueducn, h://wwwscueducn Chen J, Yin J A collaboraive filering recommendaion algorihm based on influence ses Journal of Sofware, 007,18(7:1685 1694 h://wwwosorgcn/1000-985/18/1685hm Absrac: The radiional user-based collaboraive filering (CF algorihms ofen suffer from wo imoran roblems: Scalabiliy and sarsiy because of is memory-based neares neighbor uery algorihm Iem-Based CF algorihms have been designed o deal wih he scalabiliy roblems associaed wih user-based CF aroaches wihou sacrificing recommendaion or redicion accuracy However, iem-based CF algorihms sill suffer from he daa sarsiy roblems This aer resens a CF recommendaion algorihm, named CFBIS (collaboraive filering based on influence ses, which is based on he conce of influence se and is a ho oic in informaion rerieval sysem Moreover, i defines a new redicion comuaion mehod for his new recommendaion mechanism Exerimenal resuls show ha he algorihm can achieve beer redicion accuracy han radiional iem-based CF algorihms Furhermore, he algorihm can alleviae he daase sarsiy roblem Key words: : E-commerce; recommendaion sysem; collaboraive filering; influence se,,, (, CFBIS(collaboraive filering based on influence ses, Suored by he Naional Naural Science Foundaion of China under Gran Nos60573097, 6067306 ( ; he Research Foundaion of Naional Science and Technology Plan Proec of China under Gran No004BA71A0 ( ; he Research Foundaion of Discilines Leading o Docorae Degree of Chinese Universiies under Gran No0050558017 ( ; he Naural Science Foundaion of Guangdong Province of China under Gran Nos050030, 0430046 ( ; he Research Foundaion of Science and Technology Plan Proec in Guangdong Province of China under Gran No005B1010103 ( ; he Naural Science Foundaion of Souh China Universiy of Technology under Gran NoB07E506050 ( Received 006-03-0; Acceed 006-07-05

1686 Journal of Sofware Vol18, No7, July 007,, : ; ; : TP393 : A eb, eb, eb, eb,eb eb eb eb eb,,,, (recommendaion sysem,, [1] [,3] [4] [5], (collaboraive filering 1,,,, N (o N recommendaion, (user-based collaboraive filering, :PoceLens( GrouLens [6] Usene ;TV Scou [7] eb ;Memior [8] ;Push!Music [9], : Good [10] ;Jin [11] ; Sarwar [1],, : 1 [1], NN,, [13],, ;,,,,,, 3 (new iem roblem [14],,,, (iem-based collaboraive filering [15], O(n,,n :a, ;b

: 1687,,, [16], CFBIS(collaboraive filering based on influence ses,, CFBIS,,,,, m U={u 1,u,,u m } n I={i 1,i,,i n } m n A(m,n, 1 Table 1 1 A(m n user-iem raings marix - A(m,n i 1 i i n u 1 1,1 1, / u,1 /,n u m / m, m,n, m m,n n u I I,, u, u 1 i ={ 1,,,,, m, } i ={ 1,,,,, m, }, i i i,i [17] : 1 sim i,, = 1 (, i = cos( i, i =, m m (, (, = 1 = 1,, u i,, u i m, A(m,n i i, i i, U ( u U, (, i, i =, ( ( u U, u U,,, u, 3 Pearson, U u U u U, ( u U, (, i, i =, ( (,,,, i i,,, u I u

1688 Journal of Sofware Vol18, No7, July 007 u a, : (1 u a i I, ; u a ( N, N : I r I,I r = I ua, i : ( = 1 = = 1 i, i i, i, u a i, CF 3 CFBIS (,,,,,,,, (, 31 -,,-, S -, NN RNN(reverse neares neighbor [16] d S, S =n,,d(,,,,d(, Euclidean,,D(,, RNN(={ S NN(},NN RNN, NN( / RNN(, 1 r P a, i (1 S u Fig1 The relaionshis beween NN and RNN 1 NN RNN,NN(={s,},RNN(={r,,s,u}RNN 0 (

: 1689 3, CF,, CFBIS, :,, i a i b -, i a i b (relaed; i a i b -, i b i a - ( i 1 i, (,i a i b i b i a, i i 1 i i 3 i 4 Fig A simle samle of similariy relaionshis among iems CFBIS A(m,n,,, i NN(i ={i 1 }, RNN(i ={i 1,i 3,i 4 }, u a ( i, (1 i,i 0,, 3 3 Table Iem similariy able Table 3 If u a has no raings for all neares neighbor of i, hen radiional iem-based CF aroaches canno roduce redicion raings for i =1 = =3 i 1 i i 3 i 4 i i 1 i 3 i 4 i 3 i i 1 i 4 i 4 i i 1 i 3 3 u a i i 1, CF i i 1 i i 3 i 4 u? 4 5 i 1 X?? i 3 i i 4 Fig3 i lacs necessary raings for is neighbors 3 i, i, i 3 i 4 i,

1690 Journal of Sofware Vol18, No7, July 007 33, u a i, CFBIS, i NN(i R NN(i u a i 4 : P a, i ( i NN ( i i, i + i NN ( i i NN ( i = (1 i, i + i, i ( i NN ( i i, u a i,i,i i i, NN(i R NN(i i NN ( i i NN ( i (( i, i i NN ( i i NN ( i i NN ( i = + + ( i, i + i, i i, i + i, i,, i,i i i NN ( i ( i NN ( i i, i i NN ( i (( i NN ( i, i i NN ( i I ua i i NN ( i i = α + (1 α (3 i, i + i, i i, i + i, i (, i, i i NN ( i,α, α=1,, ; α=0,,α [0,1] i NN ( i i NN ( i (( i, i i NN ( i i NN ( i = α + (1 α (4 i, i + i, i i, i + i, i 34 1 Find_SimLis : - A(m,n; i NN ( i : - T NN - T RNN (( (1 A(m,n, ; i i NN ( i ( i I, i, - T NN ; (3 T NN, i I, - T RNN Find_SimLis, (1 ( CF, (3, -, ( CF O(n ( [15] 33, Find_SimLis O(n,,,, - -, CFBIS : u a, i,,,α; :u a i P (1 - T NN i, {i 1,i,,i };, i

: 1691 ( - T RNN i, {, i,, i } ; i 1 (3 (1~(4, u a i P CFBIS, (1 (3 CF, (, R NN, (1 CF O(1(, O(1 4 PC(Penium 4,CPU 4GHz, 51M, indows XP, Java 41 MovieLens Minnesoa GrouLens, eb 43 000, 3 500, 943 1 68 100 000,, 0 1~5,5 erfec, 1 bad,, x,,x=08 80%, 0%, x=08,,, ψ, 100000 ψ = 1 = 093695 943 168, 4 MAE(mean absolue error MAE MAE, [17] MAE, N { 1,,, N }, {r 1,r,,r N MAE 43 MAE = N i = 1 i ri N, 10, 50,100,150,00,50 300, 4 4,, MAE,, CFBIS

169 Journal of Sofware Vol18, No7, July 007 14 1 10 Comarison of similariy measure mehods Cosin Adused cosin Correlaion MAE 08 06 04 0 00 10 50 100 150 00 50 300 Number of neighbors Fig4 Comarison of similariy measure mehods 44 4 5, 5, 40, ;,, (3 α, α=1,, ( [15] ; α=0, α [0,1], 01 1, (,, MAE =70;, MAE =300 =, α,, 6, (, MAE (α=1 MAE, 6, (α=05,mae MAE 093 089 085 081 Comarison of he effec of differen iemses Cosin (NN_based Cosin (RNN_based Adused cosin (NN_based Adused cosin (RNN_based MAE 077 076 075 074 Comarisons of MAE on differen alha Cosin Adused cosin 077 073 073 0 30 40 50 60 70 80 90 100 10 140 160 180 00 Number of neighbors 07 1 09 08 07 06 05 04 03 0 01 0 Alha Fig5 Comarison of he effec of differen algorihms Fig6 Comarison of MAE on differen alha which oally based NN or RNN (Formula (1 (Formula (3 5 ( (1 6 alha MAE ( (3

: 1693 45 CFBIS CFBIS ( 7 ( 8, (1, NN_based;CFBIS (1~(4, CFBIS1, CFBIS,CFBIS3 CFBIS4, (3 (4,α=05 MAE 0776 0766 0756 0746 0736 076 0716 0706 0696 0 Comarison of he NN-based algorihm and CFBIS 30 NN_based CFBIS3 40 CFBIS1 CFBIS4 CFBIS 50 60 70 80 90 10010 140160 18000 Number of neighbors Fig7 Comarison of he NN-based algorihm and CFBISs (sandard cosine similariy measure MAE 093 089 085 081 077 073 Comarison of he NN-based algorihm and CFBIS NN_based CFBIS3 CFBIS1 CFBIS4 CFBIS 0 30 40 50 60 70 80 90 10010 140160 18000 Number of neighbors Fig8 Comarison of he NN-based algorihm and CFBISs (adused cosine similariy measure 7 CFBIS ( 8 CFBIS ( 7 8, CFBIS MAE, 4,,,,, 5 eb,,, CFBIS, CF, CFBIS,,CFBIS,,,,,,,, -, - [18],, References: [1] Broadvision h://wwwbroadvisioncom [] Nanooulos A, Kasaros D, Manolooulos Y A daa mining algorihm for generalized web refeching IEEE Trans on Knowledge and Daa Engineering, 003,15(5:1155 1169

1694 Journal of Sofware Vol18, No7, July 007 [3] ang S, Gao, Li JT Real ime ersona1izaion based on classificaion Chinese Journal of Comuers, 00,5(8:845 85 (in Chinese wih English absrac [4] Jin X, Zhou Y, Mobasher B A unified aroach o ersonalizaion based on robabilisic laen semanic models of eb usage and conen In: Proc of he AAAI 004 orsho on Semanic eb Personalizaion (SP 004 San Jose: AAAI, 004 6 34 h://mayacsdeauledu/~mobasher/cgi-bin/view-ubsl?cid=um [5] Herlocer J, Konsan J, Riedl J Exlaining collaboraive filering recommendaions In: Proc of he ACM 000 Conf on Comuer Suored Cooeraive or 000 41 50 h://oralacmorg/ciaioncfm?doid=358916358995 [6] Miller B, Konsan J, Terveen L, Riedl J PoceLens: Towards a ersonal recommender sysem ACM Trans on Informaion Sysems, 004,(3:437 476 [7] Baudisch P, Bruecner L TV scou: Guiding users from rined TV rogram guides o ersonalized TV recommendaion In: Proc of he nd orsho on Personalizaion in Fuure TV Malag 00 157 166 h://wwwaricbaudischcom/ublicaions/ 00-Baudisch-TV0-TVScouGuidingUsersdf [8] DeRoure D, Hall, Reich S, Hill G, Pirais A, Sairmand M MEMOIR An oen framewor for enhanced navigaion of disribued informaion Informaion Processing and Managemen Journal (Elsevier Science, 001,37(1:53 74 [9] Holmuis LE, Jacobsson M, Ros M hen media ges wise: Collaboraive filering wih mobile media agens In: Proc of he IUI 006, he 10h In l Conf on Inelligen User Inerfaces Sydney, 006 h://oralacmorg/ [10] Good N, Schafer JB, Konsan JA, Borchers A, Sarwar BM, Herlocer J, Riedl JT Combining collaboraive filering wih ersonal Agens for beer recommendaions In: Proc of he 16h Naional Conf on Arificial Inelligence (AAAI 99 Menlo Par: American Associaion for Arificial Inelligence, 1999 439 446 h://oralacmorg/ciaioncfm?id=31514931535&coll= &dl=&cfid= 15151515&CFTOKEN=6184618 [11] Jin X, Zhou YZ, Mobasher B A maximum enroy eb recommendaion sysem: Combining collaboraive and conen feaures In: Proc of he ACM SIGKDD Conf on Knowledge Discovery and Daa Mining (KDD 005 Chicago, 005 61 617 h://oralacmorg/ciaioncfm?id=1081945&dl=&coll=&cfid=15151515&cftoken=6184618 [1] Sarwar B, Karyis G, Konsan J, Riedl J Alicaion of dimensionaliy reducion in recommender sysems A case sudy In: Proc of he ebkdd 000 orsho a he ACM-SIGKDD Conf on Knowledge Discovery in Daabases (KDD 000 000 h://cieseerissuedu/sarwar00alicaionhml [13] Deng AL, Zhu YY, Shi BL A collaboraive filering recommendaion algorihm based on iem raing redicion Journal of Sofware, 003,14(9:161 168 (in Chinese wih English absrac h://wwwosorgcn/1000-985/14/161hm [14] Mobasher B, Jin X, Zhou YZ Semanically enhanced collaboraive filering on he eb In: Berend B, e al, eds eb Mining: From eb o Semanic eb LNAI 309, Sringer-Verlag, 004 57 76 [15] Sarwar B, Karyis G, Konsan J, Riedl J Iem-Based collaboraive filering recommendaion algorihms In: Proc of he 10h In l orld ide eb Conf New Yor: ACM Press, 001 85 95 [16] Korn F, Muhurishnan S Influence ses based on reverse neares neighbor ueries In: Naughon JF, Bernsein PA, eds Proc of he ACM SIGMOD In l Conf on Managemen of Daa New Yor: ACM Press, 000 01 1 [17] Herlocer J, Konsan J, Terveen L, Riedl J Evaluaing collaboraive filering recommender sysems ACM Trans on Informaion Sysems (TOIS, 004,(1:5 53 [18] Chen J, Yin J, Chen L Research on influence ses and is dynamic indexing srucure and uery algorihm based on muli-dimensional vecors Journal of Comuer Research and Develomen, 004,41(Sul:90 95 : [3],, eb,00,5(8:845 85 [13],,,003,14(9:161 168 h://wwwosorgcn/ 1000-985/14/161hm [18],,,004,41( :90 95 (1977,,,,, eb, (1968,,,,,CCF,,