The AGA Evaluating Model of Customer Loyalty Based on E-commerce Environment

Size: px
Start display at page:

Download "The AGA Evaluating Model of Customer Loyalty Based on E-commerce Environment"

Transcription

1 6 JOURNAL OF SOFTWARE, VOL. 4, NO. 3, MAY 009 The AGA Evaluating Model of Custoer Loyalty Based on E-coerce Environent Shaoei Yang Econoics and Manageent Departent, North China Electric Power University, Baoding City, China Eail: Rui Zhang College of Inforation Science & Technology, Agricultural University of Hebei, Baoding City, China Eail: Zhibin Liu Econoics and Manageent Departent, North China Electric Power University, Baoding City, China Eail: Abstract With the developent of e-coerce, the enterprises should build long-ter and stable relationship with the custoers, and then enhance custoer continuously through the use of inforation technology and network technology, which based on the custoer interest, eet the needs of clients and custoers to create value for the as the goal. At the sae tie, for e-coerce beyond the liitations of tie and space, so that the aterial, financial and inforation can high-speed flow, the subsyste, such as guide, ordering, payent, trade and security can be linked together organically, and then achieve the goods on-line trading. So face to the coplex process and the huge inforation, how to obtain and retain ore valuable custoers and then establish and consolidate custoer, which is the key for the enterprises strategy. Based on the analysis of the custoer and probles which faced, this paper established an evaluation index syste, including five aspects: custoer expectations, custoer trust, custoer satisfaction, custoer awareness value and transfer cost; this paper built custoer odel in e-coerce environent based on introducing the principles of GA ethod and iproved progra; the exaples deonstrated that AGA ethod is scientific and operational on the issue of custoer evaluation, and the AGA ethod can be applied to further areas; finally, this paper put forward the strategies of training and consolidating custoer. Index Ters custoer, e-coerce environent, GA, AGA, evaluation odel I. INTRODUCTION In the era of network rapid developent, soe copanies have also entered the fanatical wave of building internet coerce; they have focused on how to attract custoers and then ignored how to retain the loyal custoers. In fact, in the network era, custoer is especially iportant due to the uniqueness of e- coerce. Because loyal custoer is a source of enterprises obtaining copetitive advantage, who tend to purchase repeatedly products of this website and is not sensitive to price, but also pass a good reputation initiatively for Web site, recoend new custoers, use this website products ust as in the past and not change the trust of the site for the influence of the outside world. According to the findings of the United States agencies, custoer increase 5%, corporate profits increase 5%-85%. Therefore, enterprises need a way to evaluate whether their custoers loyal, segent custoer and ipleent the different arketing strategies, so that is ore conducive to the developent of e-coerce sites. A. Custoer Loyalty Custoer is that the custoers lock in your copany and use your product ever since a long tie ago, and will choose still your copany while purchase a siilar product the next tie. We understand custoer in two ways: Attitude and tropis: The attitude and tropis is on behalf of the custoer-to-business products active-oriented degree, also reflects wishes for the custoer recoending the products to other custoers. Custoer is that for the enterprises arketing or brand personality coincide with the consuer lifestyle or values, the consuers produce sentient to the enterprises or the brand, even proud, and look it as theselves spiritual sustenance, and then show the desire of purchasing continuously. Conduct repeat: Conduct repeat is the possibility of the custoers purchasing continuously the products in an enterprise, which can be easured by the indicators, such as ratio, order and possibility of the custoers purchasing products etc. The continuous purchase behavior ay coe fro the favorable ipression to the enterprise products, ay also coe fro the factors having nothing to do with feeling, such as the purchase ipulse, the enterprise sales prootion activities, the custoer purchase habits, the too high transfer cost, the custoers can t purchase other products for the enterprise s arket 009 ACADEMY PUBLISHER

2 JOURNAL OF SOFTWARE, VOL. 4, NO. 3, MAY doinance, or purchase other products inconveniently, and so on. B. The Custoer Loyalty Issues Under the Environent of E-coerce The operation ode of e-coerce based on the inforation technology and network technology, which broken through the traditional corporate business odel, and led to the iportant changes of the copetition rules aong the enterprises. Using the low-cost and the network rapid spread, we can acquire custoers requireent, strengthen ties with custoers, excavate and anage effectively the custoers resources, attract and retain the custoers, and then obtain the arket copetitive advantage, but having soe issues of aintaining the relations with the custoers in the virtual environent, as the following: ) The lower inforation anageent level of business-to-custoer The businesses get the custoer inforation through soe ways, such as the ebership registration, the questionnaire survey, etc; draw the useful custoer behavior ode and a variety of potential inforation fro large custoer data inforation, treat differently the different custoers, provide the favorable value services to the value custoers, and then seize arket opportunities, lay the foundation for the establishent of custoer. While ost enterprises have already established a background anageent syste to safeguard custoer basic data, but which is only as to retain custoer inforation and failed to collect custoer data for statistical analysis, can not get the evaluation and recoendations of custoer-to-business-product, and even know noting about the custoers acknowledge value. In addition, due to involving the custoers privacy in the personal data, the custoers ay provide false inforation; the enterprises lack the ability to identify, which resulting in a lack of authenticity of the custoer data. As a result, enterprises can not identify and classify accurately based on the historical data and trading patterns of custoers, and then lose the best counicating tie with the valuable custoers. ) Custoer-to-business lack of trust due to the iperfect echanis In the e-coerce arket, because the econoy and legal syste can t catch up with the developent pace of e-coerce far and away at present in China, the e- coerce transactions ways bring the separation of tie and space between buyers and sellers, and then produce the trust and confidence difficulties. The reason is that the traditional reliance and confidence basis is weakened due to the virtual anonyity of e-coerce, a large-scale opening inforation infrastructure are very fragile facing to coputer crie and network fraud, and e-coerce security can not obtain a good guarantee. As the difficulteasy degree of confidence-building between people has soething to do with their trading ethods, in e- coerce the custoers often understand product only through the pictures or character description, the inforation asyetry issue of product or service quality is ore serious as a result of can not or difficulty to observe in the e-coerce arket, so it is ore difficult that a business wins the custoers trust. 3) Enterprises lack the sense of establishing personalized service for the needs of different custoers In the e-coerce environent, custoers will be able to ski over a large nuber of product inforation never leaving their hoe, which requires businesses Web sites to establish corresponding and effective counication ode and prootional content based on the consuers different preferences, interests and deands, in order to coplete one-on-one service in the low-cost, which requires copanies to analyze the deands of the different custoers. "Tailored" can eet the requireents, which can not only reduce the tie of skiing over the product inforation based on the custoers spend oney, but also enhance the custoers satisfaction, and bring the loyal and stable custoer base for businesses. However, the vast aority of enterprises istake e-coerce for the networking of business activities akes the corporate Web site as the window of inforation disseination; there is no use of the advantages of e-coerce to eet the custoers different deands, which can not aintain custoer. At present, there are gaps in training and consolidating custoer ethods for China enterprises in e- coerce environent. For a long tie, China enterprises attach iportance to qualitative analysis in custoer evaluation; furtherore, these analysis ethods are not coprehensive. It is indispensable to strengthening custoer anageent, but with foreign custoer anageent ethods, which is lack of quantitative analysis and still iprecise in custoer identification, easureent and other aspects. In addition, the biggest obstacle is custoer inforation syste construction seriously lagged behind in China enterprises iproving custoer anageent ethods in e-coerce environent. For the large nuber of business and custoers inforation deficiency, the enterprises unable to establish the corresponding anageent odel and can t grasp accurately the custoers exposure. Custoer inforation is distortion, which ipact directly the custoer anageent decision-aking scientificalness, and add ore difficult to custoer evaluation ethods quantitative. In view of this, this paper constructs scientific evaluation index syste, including five aspects: custoer expectations, custoer trust, custoer satisfaction, custoer awareness value and transfer cost, and solves lack of quantitative analysis proble through applying the Genetic Algorith (GA), Adaptive Genetic Algorith (AGA) and iproved Genetic Algorith. II. THE CUSTOMER LOYALTY EVALUATION INDEX SYSTEM CONSTRUCTION IN THE E-COMMERCE ENVIRONMENT 009 ACADEMY PUBLISHER

3 64 JOURNAL OF SOFTWARE, VOL. 4, NO. 3, MAY 009 A. The Characteristics of Custoer Loyalty in E- coerce Environent Fro the e-coerce characteristics, due to the use of the Internet, the establishing process of the custoer is different fro the traditional custoer, the ain characteristics show in the following areas: ) Custoers recoending is faster and broader In the past custoers recoending was through telephone or face-to-face anner, so it is slow and not wide range; in the age of e-coerce, the Internet as an ideal counication channel, which will enable the custoers happy consuer experience to be shared ore quickly and widely due to the characteristic of the rapid spread, thereby culture and create greater opportunities to the custoers. ) Enterprises can provide greater value to custoers In the condition of Internet, the custoers deands can be better satisfied, such as the personalized deand, fast processing, etc, which is very difficult to achieve in the past. 3) Easier to build relationships of business-tocustoer The Internet is an interactive counication tool, which can not only facilitate the custoers feedback to the enterprise conduct, but also facilitate the enterprises to pay close attention to each custoer, therefore, the excellent relations between the enterprise and custoers is easier to establish. In contrast, the traditional counication channels, such the opinion letters, service phone, custoers receive, etc, which are on pales beside. 4) The cost which enterprise acquire custoers is greater, but the profit is higher The virtual nature of the Internet led to uncertainty and risk of bargaining, so the enterprises ust pay a uch higher cost when build good relationship with the custoers in early days. However, the enterprises use network tools to provide greater added value, so the custoers have easily, along with the enterprises profits will be greater. B. The Custoer Loyalty Evaluation Index Syste ) Custoer expectations (B ) In the e-coerce environent, custoers can understand all business inforation in detail through network platfor easily, and have an advance look forward to products or services of their interest; custoers deands further iprove the specialization, personalization, convenience, quick response of product or service and so on, which are particular characteristics relative to the traditional business environent. Therefore, in the e-coerce environent should be an additional evaluation: the recoendation or word-of-outh of the edia and other custoers (C ); the related detailed product inforation quality and quantity of web site (C ); Web site interactive (C 3 ) (tie and convenience of searching for inforation); specialization of products or services (C 4 ); personalization of products or services (C 5 ); rapid response of products or services (C 6 ). ) Custoer trust (B ) In arketing trust theory, trust is the basis of directly, it is necessary to built custoer trust in order to succeed in establishing a high level of long-ter custoer relationships, therefore trust is a decided factor of custoer. Custoer trust eans dependent willingness custoers to credible trading partners, including two diensions, which are the credibility and goodwill: the strength of eeting the deand (C ), honesty (C ) and fairness (C 3 ) and so on. In the e- coerce environent should be an additional evaluation, which are safety (C 4 ) and reliability (C 5 ) of the site (payent security, privacy protection, security policies, credit syste and legal environent, etc). 3) Custoer satisfaction (B 3 ) Satisfaction is a feeling state level, which coes fro the coparison between perception perforance or output for products or services and the expectations. Custoer satisfaction is ainly the total of post-arket evaluation supplier to existing custoers: satisfaction to sales (C 3 ); satisfaction to the staff of after-sales, technical supporting and training (C 3 ); eotional factors (such as pleasure feelings) (C 33 ). In the e-coerce environent should be an additional evaluation, which is on-line services (C 34 ) (advice, help, application, registration, search and change, and so on). 4) Custoer awareness value (B 4 ) Custoer awareness value is subective evaluation of custoers-to-supplier relative value, including: product features (C 4 ); product price(c 4 ); product quality(c 43 ); product branding(c 44 ); custoer service(c 45 ) and other invisible costs(c 46 ) (conversion cost, use cost, tie costs, spirit costs, physical strength costs, etc.). In the e- coerce environent, although the custoers felt a large nuber of convenience due to the network, but also felt a lot of risk, therefore, should also be an additional evaluation of custoer perception risk: products risk (C 47 ) (the custoers can t experience and check personally product quality, and also can not distinguish the products types); security risk (C 48 ) (arising fro the network trading, exchange data in an open networks, be likely to daage or leakage privacy). 5) Transfer cost (B 5 ) Transfer cost refers to the related costs, involve which the custoers and existing suppliers end the relationship, and establish a new replaced relationship, in the process of aintaining relations between the custoers and forer corporate, which is the results of tie, energy, knowledge, feelings and the physical capital investing in products, services and relationship, including the relationship interests (C 5 ) (The special interests of custoers due to relationships of business and custoers, the copetitors can not provide custoers with the sae interests in short order after custoers conversion); the resource costs (C 5 ) (Including the cost of terinating service with existing provider and resources loss of transact with the new supplier, those, including oney, tie, energy and the property cost); the psychological costs (C 53 ) (Custoer psychological pressure when convert product or service provider, including interpersonal conflicts and eotional risk-aware 009 ACADEMY PUBLISHER

4 JOURNAL OF SOFTWARE, VOL. 4, NO. 3, MAY conversion); alternative liit (C 54 ) (the nuber of product or service providers who custoers can choose ay also affect custoer conversion behavior, which is obective obstacles while the custoer conversion); additional services or service resuption (C 55 ) (including professional services for the custoer personalized deand, all the actions and efforts of service organizations in order to ake up for custoer loss after the experience of services failure, the good service resuption can create value for custoers and prevent effectively custoers loss because of service failure). In the e-coerce environent, due to a sharp drop in corporate onopoly and the custoer search costs, we should focus on evaluation of psychological costs and services costs in order to reduce the custoer loss trends. III. THE GA MODEL CONSTRUCTION A. The Basic Principle of GA Genetic Algorith cae of the coputer siulation research to the biology systes. Professor Holland in Michigan University inspired by the biology siulation technology, created a self-adaptation probability optiization technology which fit for the coplex syste optiization based on the biology genetic and evolutionary echanis, that is the genetic algorith. Copared with the genetic algorith, the ost classical optiization algorith is the gradient or higher tie statistics based on a single easuring function (evaluating function) to produce a deterinate experientation solution sequence; Genetic algorith is not dependent on gradient inforation, but search for the optial solution through siulated natural evolutionary process. It uses the coding technology to act on the nuber bunch called chroosoe, siulates the evolutionary process that coposed of these nuber bunches. Genetic algorith regroups the good adaptability bunches, and generates the new bunch groups through the organized and rando inforation exchange. B. The Model of GA ) The indexes standardization To the custoer probles, we suppose the evaluating set A= {A l, A,,A n }, the index set G= {G l, G,, G }, the index value of proble A i to index G is x i (i=,,, n; =,,, ). The coprehensive evaluating indexes of custoer probles include custoer expectations, custoer trust, custoer satisfaction, custoer awareness value and transfer cost. Suppose the vector Q l ={Gl, G,,G k }denotes k custoer expectations index, the vector Q ={G k+l, G k+,,g p }denotes p-k custoer trust and satisfaction index, the vector Q 3 ={G p+, G p+,,g q }denotes q-p custoer awareness value index, the vector Q 4 ={G q +l, G q+,,g }denotes -q transfer cost index. This paper carries through diensionless disposal to the probles used fuzzy subection function, the result is as follows: X= {x i =,,,n; k=,,,} For custoer expectations index, its subection function is: xi a b xi () y = a < x i i < b b a xi b 0 (i=,,, n); Q x i denotes the index value, y i denotes the value after diensionless disposal, a and b denote the axiu and the iniu of the indexes. For custoer trust and satisfaction index, its subection function is: xi b x a () i y = a < x i i < b b a xi a 0 (i=,,,n); Q For custoer awareness value index, its subection function is: q xi xi < q in ax ax{ q, x, x q} (3) yi = q xi q xi q in ax ax{ q, x, x ( k) q} xi > q (i=,,,n); Q 3 For transfer cost index, its subection function is: xi = x (4) y = xi x i ax xi x x i x i (i=,,,n); Q 4 In the forula, x ax and x in denote the axiu and iniu value of transfer cost index respectively. ) Deterination of the obective function Suppose the benchark proect Ao={x 0 =,,, ), the ethod to select the benchark proect is as follows: To the custoer expectations index, order: x o = in{ x i } i n Q (5) To the custoer trust and satisfaction index, order: x o = ax{ x i } i n Q (6) To the custoer awareness value index, suppose the [q l, q ] is the optial custoer awareness value of the index value, order: xo = x ( x [ q, q ]) Q3 (7) To the transfer cost index, suppose the x is the optial index fixed value, order: x o = x Q 4 (8) Suppose the W= {w l,w,...,w } is the index weight value, thereinto (=,,...,), and W eets the constraint condition: w = = (9) 009 ACADEMY PUBLISHER

5 66 JOURNAL OF SOFTWARE, VOL. 4, NO. 3, MAY 009 The index value of the benchark proect A 0 After the standardization is y 0 = {y o =,,, }, w y 0 denotes the weighted coprehensive perforance value of the benchark proect index, and the w y i denotes the weighted coprehensive perforance value of the proect i. The change of the weighted coprehensive perforance values between the proect index and the benchark proect index should be stable relatively, naely, the saller the deviation is, the better the proect is. Suppose Z i is the weighted coprehensive perforance value deviation between proect i and the benchark proect, the obective function can be given under the constraint condition (9): in Zi ( W ) = w = = w ε ( w y0 w yi ) = ( =,,... n) ( 0 < ε < ; =, ),..., (0) In the forula (0), ε is the stated iniu value of the index weighted vector, in order to solve the above proble, we can assebles the obective function as the obective function U(W) through the tantaount weight disposal: n n U( W ) = Zi ( W ) = w ( y0 yi ) () i= n i= n = The forula () divided by n on both sides of the equation, suppose F (W) = U(W)/n, we can gain: F( W ) = w ( y0 yi ) = w ( y0 y ) = n = n y = yi i= n () (=,,,) Then forula (0) can transfer into forula (3), naely, the obective function is: in F( W ) = w ( y0 y ) = w = s. t. = w ε (0<ε <; =,,,) IV. THE AGA AND IMPROVED GA MODEL (3) A. The AGA Basic Principle The GA including proportion copy, adaptive exchange and utation operation is known as the AGA. During the searching for the optiizing paraeters, AGA can aintain the solution groups diversity and convergence function through changing adaptively the exchange probability P C and utation probability P based on the solution groups environent adaptive ability. AGA can control adaptively the search process and then get the global optial solution through acquiring and accuulating autoatically the knowledge about space search in the search process. And prove that AGA has the global convergence by use of Markov chain, that is, AGA can converge to the optial solution. B. Adaptive Exchange Probability and Mutation Probability Exchange operation and utation operation play an iportant role in GA. Exchange operation plays a aor role, which cobines and exchanges the valuable inforation between the two individual, generates new future generations, and then accelerates the search speed greatly during the groups evolution; Mutation operation plays an assistant role, which is accidental and secondary, can aintain the groups gene diversity. In the course of specific operations, the AGA can change exchange probability P C and utation probability P, aintain the diversity, prevent preature convergence, and then increase the calculation ethod speed and accuracy based on the individual specific circustances. It is achieved that change adaptively P C and P through inspecting the relationship between the groups average fitness value f avg and the largest fitness value f ax. For siple genetic algorith (SGA), under noral circustances, when it reached a superior solution, the group ove closer to the superior solution quickly, this can reduce the individual fitness differences, that is, f ax - f avg lower. With the f ax -f avg change, the forula of P C and P are: P C = k /( f ax f avg ) (4) P = k f f ) ( ax avg (5) P C and P don t depend on any solution individual's fitness value, for all groups, P C and P have the sae value. High fitness degree solution individual and low fitness degree solution individual have the sae exchange frequency and utation frequency. When the groups converge to the optial solution, P C and P increase, and is likely to cause the high perforance solution individual being destroyed in the vicinity of the optial solution. In order to overcoe these probles, it is necessary to protect the good perforance solution in the group. For the high fitness degree solution individual, we use the lower exchange probability P C and utation probability P to ensure the GA convergence; for the low fitness solution individual, we use the higher exchange probability P C and utation probability P to prevent SGA preature convergence. Therefore, P C depends not only on f ax- f avg, but also on the two series of fitness value used for exchange. Siilarly, P depends not only on f ax -f avg, but also on the fitness value of the pending utation individual. The expressions of P C and P are: ' ' k ( f ax f )( f ax f avg ) if f > f avg PC = ' k3 if f < f avg (6) k ( fax f )( fax favg) if f > favg P = k4 if f < favg (7) In the forula, f ax is the largest fitness degree of the group; f avg is the average fitness degree of the group; f is 009 ACADEMY PUBLISHER

6 JOURNAL OF SOFTWARE, VOL. 4, NO. 3, MAY the greater fitness degree of the two exchange series; f is the fitness degree value of the pending utation individual. General recoended value, k =k 3 =, k =k 4 =0.5. The larger the P C, the ore new individual could be introduced in groups, but the too uch exchange operation of P C could lead to the groups non-evolutionary, because its speed ay be faster in ters of underining the high-perforance character string structure than producing the new gifted individual. The saller the P C the lower speed of searching for new individual could lead to the stagnant. Exchange operation plays a aor role in GA algorith, which can increase greatly the search speed during the group evolution, so the value of P C is often taken between 0.5 and.0. Mutation is a secondary operator, which is ainly used to increase the group changes. The low utation rate can prevent the any one to reain the sae forever during the search process, and guarantee the algorith can search every point of the proble space; but the high utation rate incline to search iediately. So the value of P is saller, fro to In the practical application process, we ust adust appropriately k and k to ensure the value of P C and P in the certain reasonable liits. C. Solving with the iproved genetic algorith For the above non-linear prograing proble, this paper solves with the iproved genetic algorith. The aor steps are as follows:. Generate an initial solution group randoly. Suppose the initial solution group U l ={x l }, ( =,,,), t denotes the evolutionary generation.. Calculate the fitness function value of current solution. According to the fitness function definitions, calculate the chroosoe (the candidate solutions) fitness function eval(x l ), (=,,,). 3. According to the proportion inforation of the chroosoe, the bigger the fitness degree of the candidate solution is, the bigger the probability of participating to generate the next generation is. 4. Select two vectors of x l i and x l with the probability P c fro the vector x l, x l,, x l i,, x l,, x l, carry through the intercrossing operations according to the new operator. Reserve a vector after the intercrossing operations randoly and discard another vector, gain the candidate solution {x l, x l,, x l i,, x l,, x +l }. 5. Select a certain vector in deterinate probability P aong the vector set after the intercrossing operations, carry through the aberrance operations according to the new aberrance operator to the vector, and gain the candidate solution set of next generation {x +l, x +l,, x +l,, x +l }. 6. Suppose the Euclidean distance of the two vector solutions that the fitness degree is highest in the candidate solutions set is: l l + d = x x = ( x ax ax t t + t t + ax x ax ) ( x ax n x ax ) If d ε,x*=x t+ ax gains the overall sallest point, and the algorith is convergent; Otherwise, transfers to step 4, continues to solution's optiize convergence search. V. APPLICATION EXAMPLES Based on the forer the index syste, in this paper, we divide the custoer degree into five grades, Ⅰ (Coplete ), Ⅱ (Severe ), Ⅲ (Moderate ), Ⅳ (Mild ) and Ⅴ (No ). According to the hidden layer nodes epirical forula and cobining training, we copare, adust and then identify that five nodes is optial in hidden layer, and the output layer has one node. Set the custoer level Ⅰ - Ⅴ corresponding respectively the custoer classification results of the network output: 0., 0.3, 0.5, 0.7, 0.9. Beiing enterprises data as a training saple, we carry out the network training of the entire custoer evaluation. When the data learning reoved fro 0 regions, the learning convergence is very slow, and even arise paralysis phenoenon. To this end, we carry out standardization pretreatent of the training saple data, suppose X ax is the ties of No standard value, X in is 0. ties of Coplete standard value, after the corresponding standardization, the variable is: X = ( X X in ) /( X ax X in ) Then the data saples will be noralization in [0., 0.9], and the network learning speed accelerate. The unbiased variance of the learning output error reach E-5, the training end, and the evaluation results of the training saples as shown in table I. Custoer Expected output Network output TABLE I. THE TRAINING RESULTS OF SAMPLE DATA Coplete Severe Moderate Mild No Selecting and onitoring five enterprises custoer indicators in Shanghai, the processing data as shown in table II. TABLE II. THE CUSTOMER LOYALTY MONITORING DATA OF SHANGHAI ENTERPRISES Evaluation factor Monitoring point C C C 3 C 4 C 5 C 6 C CONTINUE TABLE Evaluation factor Monitoring point C C 3 C 4 C 5 C 3 C 3 C ACADEMY PUBLISHER

7 68 JOURNAL OF SOFTWARE, VOL. 4, NO. 3, MAY Monitoring point CONTINUE TABLE Evaluation factor C 34 C 4 C 4 C 43 C 44 C 45 C Monitoring point CONTINUE TABLE Evaluation factor C 47 C 48 C 5 C 5 C 53 C 54 C Copute the network output results ebership, according to custoer and using principles of the ebership biggest, udge custoer rating of the different levels of ebership, so we can see clearly the grade-level, and then through the analysis of the saple evaluation factor, we can understand clearly the worst rating factors, that is No factors. We infer according to the trained evaluation network, the evaluation results as shown in Table III. TABLE III. THE CUSTOMER LOYALTY LEVEL RESULTS BASED ON AGA EVALUATION No Network output I(0.) II(0.3) III(0.5) No Network output CONTINUE TABLE IV(0.7) V(0.9) Evaluation results Ⅲ Ⅱ IV Ⅰ Ⅱ For ease of coparison and analysis, we evaluate the onitoring data using the traditional BP algorith and AGA, under the sae conditions, which the learning error of the network training odel is E-5, the convergence rate s coparison results between AGA network training odel and traditional BP odel as shown: the convergence tie of AGA is 7.68 S, and the convergence tie of traditional BP odel is 6.54 S. Clearly, AGA is not only faster than traditional BP algorith, but also can overcoe the probles of BP algorith easily into the local iniu. VI. CONCLUSIONS A. The Suary of AGA AGA is used in custoer evaluation syste in the e-coerce environent, integrate effectively the characteristics, that is, GA s global search ability is strong and gradient drop ethod s local search ability is strong, then GA s identification effect enhance. The network odel doesn t have to re-deterine the weight value of the evaluation factors, and can adust autoatically the ratio relationship of the various factors, so the evaluation results have strong obectivity. At the sae tie, Beiing enterprises custoer evaluation index data as the network operator saples, we evaluate the saples for evaluation using weight value and threshold value of coputing end, the odel s coonality is very good. The coparison of the traditional BP algorith and AGA through the onitoring data showed that, custoer evaluation ethod based on AGA is high speed and accuracy, which has highly robust, can reduce the network training nuber of ties, shorten the network training tie, siulate effectively the coplex non-linear relationship between evaluation factors and evaluation results, and provide a new effective approach to custoer coprehensive evaluation. B. The Strategies of Training Custoer Loyalty The key of cultivate long-ter custoer is to the custoers personalized values as guidance and create value for custoers; the values of different custoers ay be very different. Specifically, ainly in the following areas to cultivate a sense of custoer : ) Create ore value for custoers Custoer Value is the coparing results of the total cost and the total revenue when the custoers buy products or services. The total incoe is ore than the total cost; the custoer received value is greater. To this end, we can use the value ethod to analysis the interests and the core needs of the custoers. At the sae tie, in order to gain copetitive advantage, the enterprise ust be able to create value for its custoers ore than the copetitors. The custoers are satisfied with ore consuption value, the satisfied custoers is only likely to be loyal to a particular enterprise. ) Allow custoers to have a sense of trust Trust is a prerequisite which the custoers have a loyal. On-line, the word trust is particularly iportant. As the network virtual, the custoers and enterprises bargain under the condition of "invisible and intangible" each other, the custoers assue a great risk. As a result, the custoers tend to aintain long-ter relationship with the enterprises which they trust. In fact, while any custoers choose and evaluate on-line business, "reliable" is the ost iportant factor, rather than 009 ACADEMY PUBLISHER

8 JOURNAL OF SOFTWARE, VOL. 4, NO. 3, MAY "cheap" or "a wide range of products." Trust coes fro any aspects, such as high-quality products or services, reasonable prices, and so on. And on-line, there are crucial factors: protect the custoers online security, that is, pay on-line security and personal privacy security; perforance the contracts tiely and accurately; prevent the transactions fraud, and so on. 3) Encourage custoers to participate in virtual counities For businesses, the best custoers are those that put forward the representative needs and suggestions to the enterprises, which have advance awareness, interested in work together with enterprises, and then solve the proble ointly. It is very praiseworthy for the enterprises developent that has a group of "consultants" and "partnership" custoers who never satisfy. On the Internet, create virtual counities where all the custoers can counicate and help each other, the custoers can put forward good suggestions and views of enterprises products and services, the enterprises resolve custoers probles on the basis of these proposals, and then design ore appropriate products and services for the custoers. 4) Shape the custoer's personalized needs and shopping experience With the rising living standards, ore and ore custoers request is raising, the personalized needs turn into the trends gradually. There are different between the personalized production and constitution: Constitution is that, the producer design the products and services according to the custoer deand, the custoers select the product or service characteristics which they need fro a range of enu, and then tell the producers on-line shopping. And the personalized production is a process, which the custoers participate in product design and shape needs together with the business. In the network inforation environent, the enterprises can ake use of network technology to counicate directly with custoers and explore ointly the product design and production. According to the custoers needs, the enterprises design and product the products and services, which enable a high degree of satisfaction with the custoers. REFERENCES [] Fenhe Zhi, Ling Tian, and Jinghai Ao, Custoer Loyalty in EB [J], Coercial research, 007,, pp [] Xiaoni Dong and Guangrui Wen, The Strategy Analysis of Bringing up Custoer Loyalty Degree under the Environent of E-coerce [J], Shandong Spinning Econoy, 008, 4, pp [3] Jing Ni, Guangle Yan, Liangwei Zhong and Xiaoli Zhang, Research on Strategies of prooting Custoer Loyalty in EB [J], Industry Technology and Econoy, 006, 4, pp [4] Yaing Sun, Research on Custoer Loyalty in EB [J], Neiiang Science and Technology, 007, 0, pp [5] Lin Wang and Yurong Zeng, The Custoer Loyalty Manageent in EB [J], Science and Technology Manageent Research, 005,, pp [6] Jingwen Tian, Meiuan Gao, The research and application on artificial neural network algorith[m], Beiing institute of technology press, Beiing, 006,7, pp [7] Aizhen Luo, Jin Li, Evaluation of constructing indoor air quality based on iproving genetic calculation [J], Shanxi Architecture, Jul. 007, pp: [8] Ling Zhang, Yong Liu, Wei He, Application of adaptive genetic algorith in license plate location [J], Coputer Applications, Jan. 008, pp: [9] W.H. Ip., M. Huang, K.L. Yung, and D. Wang, Genetic Algorith Solution for a Risk-Based Partner Selection Proble in a Virtual Enterprise, Coputers & Operations Research, no. 30, 003, pp. 3-. [0] Fischer M., John H., and Teich T., Optiizing the Selection of Partners in Production Networks, Robotics and Coputer-Integrated Manufacturing, no. 0, 004, pp [] J.C. Kuang, and X.H. Chen, The Model of Evaluating Real Estate Investent Risks Based on Upgraded Anagentic Algorith, Coercial Research, no., 006, pp [] P. Su, N.Q. Wu, and Z.Q. Yu, Application of Iproved Genetic Algorith in Partner Selection and Optiization for a Virtual Enterprise, Syste Engineering Theory and Practice, no., 006, pp Shaoei Yang, was born in Handan City, China, and graduated fro the agricultural university of Hebei in 003, gained the aster's degree of anageent. The author s aor field of study is the business anageent. Since 003, she is always working at the North China Electric Power University, Baoding City, China. And she has published ore than papers and book. Such as the Electric Power Enterprise Manageent (Beiing: Chinese Electric Power Publishing Copany). 009 ACADEMY PUBLISHER

PERFORMANCE METRICS FOR THE IT SERVICES PORTFOLIO

PERFORMANCE METRICS FOR THE IT SERVICES PORTFOLIO Bulletin of the Transilvania University of Braşov Series I: Engineering Sciences Vol. 4 (53) No. - 0 PERFORMANCE METRICS FOR THE IT SERVICES PORTFOLIO V. CAZACU I. SZÉKELY F. SANDU 3 T. BĂLAN Abstract:

More information

Research Article Performance Evaluation of Human Resource Outsourcing in Food Processing Enterprises

Research Article Performance Evaluation of Human Resource Outsourcing in Food Processing Enterprises Advance Journal of Food Science and Technology 9(2): 964-969, 205 ISSN: 2042-4868; e-issn: 2042-4876 205 Maxwell Scientific Publication Corp. Subitted: August 0, 205 Accepted: Septeber 3, 205 Published:

More information

International Journal of Management & Information Systems First Quarter 2012 Volume 16, Number 1

International Journal of Management & Information Systems First Quarter 2012 Volume 16, Number 1 International Journal of Manageent & Inforation Systes First Quarter 2012 Volue 16, Nuber 1 Proposal And Effectiveness Of A Highly Copelling Direct Mail Method - Establishent And Deployent Of PMOS-DM Hisatoshi

More information

An Improved Decision-making Model of Human Resource Outsourcing Based on Internet Collaboration

An Improved Decision-making Model of Human Resource Outsourcing Based on Internet Collaboration International Journal of Hybrid Inforation Technology, pp. 339-350 http://dx.doi.org/10.14257/hit.2016.9.4.28 An Iproved Decision-aking Model of Huan Resource Outsourcing Based on Internet Collaboration

More information

CRM FACTORS ASSESSMENT USING ANALYTIC HIERARCHY PROCESS

CRM FACTORS ASSESSMENT USING ANALYTIC HIERARCHY PROCESS 641 CRM FACTORS ASSESSMENT USING ANALYTIC HIERARCHY PROCESS Marketa Zajarosova 1* *Ph.D. VSB - Technical University of Ostrava, THE CZECH REPUBLIC arketa.zajarosova@vsb.cz Abstract Custoer relationship

More information

CLOSED-LOOP SUPPLY CHAIN NETWORK OPTIMIZATION FOR HONG KONG CARTRIDGE RECYCLING INDUSTRY

CLOSED-LOOP SUPPLY CHAIN NETWORK OPTIMIZATION FOR HONG KONG CARTRIDGE RECYCLING INDUSTRY CLOSED-LOOP SUPPLY CHAIN NETWORK OPTIMIZATION FOR HONG KONG CARTRIDGE RECYCLING INDUSTRY Y. T. Chen Departent of Industrial and Systes Engineering Hong Kong Polytechnic University, Hong Kong yongtong.chen@connect.polyu.hk

More information

Applying Multiple Neural Networks on Large Scale Data

Applying Multiple Neural Networks on Large Scale Data 0 International Conference on Inforation and Electronics Engineering IPCSIT vol6 (0) (0) IACSIT Press, Singapore Applying Multiple Neural Networks on Large Scale Data Kritsanatt Boonkiatpong and Sukree

More information

Method of supply chain optimization in E-commerce

Method of supply chain optimization in E-commerce MPRA Munich Personal RePEc Archive Method of supply chain optiization in E-coerce Petr Suchánek and Robert Bucki Silesian University - School of Business Adinistration, The College of Inforatics and Manageent

More information

An Innovate Dynamic Load Balancing Algorithm Based on Task

An Innovate Dynamic Load Balancing Algorithm Based on Task An Innovate Dynaic Load Balancing Algorith Based on Task Classification Hong-bin Wang,,a, Zhi-yi Fang, b, Guan-nan Qu,*,c, Xiao-dan Ren,d College of Coputer Science and Technology, Jilin University, Changchun

More information

Analyzing Spatiotemporal Characteristics of Education Network Traffic with Flexible Multiscale Entropy

Analyzing Spatiotemporal Characteristics of Education Network Traffic with Flexible Multiscale Entropy Vol. 9, No. 5 (2016), pp.303-312 http://dx.doi.org/10.14257/ijgdc.2016.9.5.26 Analyzing Spatioteporal Characteristics of Education Network Traffic with Flexible Multiscale Entropy Chen Yang, Renjie Zhou

More information

Evaluating Inventory Management Performance: a Preliminary Desk-Simulation Study Based on IOC Model

Evaluating Inventory Management Performance: a Preliminary Desk-Simulation Study Based on IOC Model Evaluating Inventory Manageent Perforance: a Preliinary Desk-Siulation Study Based on IOC Model Flora Bernardel, Roberto Panizzolo, and Davide Martinazzo Abstract The focus of this study is on preliinary

More information

An Integrated Approach for Monitoring Service Level Parameters of Software-Defined Networking

An Integrated Approach for Monitoring Service Level Parameters of Software-Defined Networking International Journal of Future Generation Counication and Networking Vol. 8, No. 6 (15), pp. 197-4 http://d.doi.org/1.1457/ijfgcn.15.8.6.19 An Integrated Approach for Monitoring Service Level Paraeters

More information

Software Quality Characteristics Tested For Mobile Application Development

Software Quality Characteristics Tested For Mobile Application Development Thesis no: MGSE-2015-02 Software Quality Characteristics Tested For Mobile Application Developent Literature Review and Epirical Survey WALEED ANWAR Faculty of Coputing Blekinge Institute of Technology

More information

Use of extrapolation to forecast the working capital in the mechanical engineering companies

Use of extrapolation to forecast the working capital in the mechanical engineering companies ECONTECHMOD. AN INTERNATIONAL QUARTERLY JOURNAL 2014. Vol. 1. No. 1. 23 28 Use of extrapolation to forecast the working capital in the echanical engineering copanies A. Cherep, Y. Shvets Departent of finance

More information

Online Bagging and Boosting

Online Bagging and Boosting Abstract Bagging and boosting are two of the ost well-known enseble learning ethods due to their theoretical perforance guarantees and strong experiental results. However, these algoriths have been used

More information

The Research of Measuring Approach and Energy Efficiency for Hadoop Periodic Jobs

The Research of Measuring Approach and Energy Efficiency for Hadoop Periodic Jobs Send Orders for Reprints to reprints@benthascience.ae 206 The Open Fuels & Energy Science Journal, 2015, 8, 206-210 Open Access The Research of Measuring Approach and Energy Efficiency for Hadoop Periodic

More information

ASIC Design Project Management Supported by Multi Agent Simulation

ASIC Design Project Management Supported by Multi Agent Simulation ASIC Design Project Manageent Supported by Multi Agent Siulation Jana Blaschke, Christian Sebeke, Wolfgang Rosenstiel Abstract The coplexity of Application Specific Integrated Circuits (ASICs) is continuously

More information

Fuzzy Sets in HR Management

Fuzzy Sets in HR Management Acta Polytechnica Hungarica Vol. 8, No. 3, 2011 Fuzzy Sets in HR Manageent Blanka Zeková AXIOM SW, s.r.o., 760 01 Zlín, Czech Republic blanka.zekova@sezna.cz Jana Talašová Faculty of Science, Palacký Univerzity,

More information

SUPPORTING YOUR HIPAA COMPLIANCE EFFORTS

SUPPORTING YOUR HIPAA COMPLIANCE EFFORTS WHITE PAPER SUPPORTING YOUR HIPAA COMPLIANCE EFFORTS Quanti Solutions. Advancing HIM through Innovation HEALTHCARE SUPPORTING YOUR HIPAA COMPLIANCE EFFORTS Quanti Solutions. Advancing HIM through Innovation

More information

Fuzzy Evaluation on Network Security Based on the New Algorithm of Membership Degree Transformation M(1,2,3)

Fuzzy Evaluation on Network Security Based on the New Algorithm of Membership Degree Transformation M(1,2,3) 324 JOURNAL OF NETWORKS, VOL. 4, NO. 5, JULY 29 Fuzzy Evaluation on Networ Security Based on the New Algorith of Mebership Degree Transforation M(,2,3) Hua Jiang School of Econoics and Manageent, Hebei

More information

SOME APPLICATIONS OF FORECASTING Prof. Thomas B. Fomby Department of Economics Southern Methodist University May 2008

SOME APPLICATIONS OF FORECASTING Prof. Thomas B. Fomby Department of Economics Southern Methodist University May 2008 SOME APPLCATONS OF FORECASTNG Prof. Thoas B. Foby Departent of Econoics Southern Methodist University May 8 To deonstrate the usefulness of forecasting ethods this note discusses four applications of forecasting

More information

Extended-Horizon Analysis of Pressure Sensitivities for Leak Detection in Water Distribution Networks: Application to the Barcelona Network

Extended-Horizon Analysis of Pressure Sensitivities for Leak Detection in Water Distribution Networks: Application to the Barcelona Network 2013 European Control Conference (ECC) July 17-19, 2013, Zürich, Switzerland. Extended-Horizon Analysis of Pressure Sensitivities for Leak Detection in Water Distribution Networks: Application to the Barcelona

More information

A Case Study of Bee Algorithm for Ready Mixed Concrete Problem W. Wongthatsanekorn, N. Matheekrieangkrai

A Case Study of Bee Algorithm for Ready Mixed Concrete Problem W. Wongthatsanekorn, N. Matheekrieangkrai A Case Study of ee Algorith for Ready Mixed Concrete Proble W. Wongthatsanekorn, N. Matheekrieangkrai Abstract This research proposes ee Algorith (A) to optiize Ready Mixed Concrete (RMC) truck scheduling

More information

An Expert Decision Support System for Crisis Management. Boumediene Belkhouche Reda Bakeer Salah Al-Saleh

An Expert Decision Support System for Crisis Management. Boumediene Belkhouche Reda Bakeer Salah Al-Saleh An Expert Decision Support Syste for Crisis Manageent Bouediene Belkhouche Reda Bakeer Salah Al-Saleh EECS Departent Tulane University New Orleans, LA 70118 (504) 862-3372 (phone) (504) 862-3293 (fax)

More information

This paper studies a rental firm that offers reusable products to price- and quality-of-service sensitive

This paper studies a rental firm that offers reusable products to price- and quality-of-service sensitive MANUFACTURING & SERVICE OPERATIONS MANAGEMENT Vol., No. 3, Suer 28, pp. 429 447 issn 523-464 eissn 526-5498 8 3 429 infors doi.287/so.7.8 28 INFORMS INFORMS holds copyright to this article and distributed

More information

An Approach to Combating Free-riding in Peer-to-Peer Networks

An Approach to Combating Free-riding in Peer-to-Peer Networks An Approach to Cobating Free-riding in Peer-to-Peer Networks Victor Ponce, Jie Wu, and Xiuqi Li Departent of Coputer Science and Engineering Florida Atlantic University Boca Raton, FL 33431 April 7, 2008

More information

Investing in corporate bonds?

Investing in corporate bonds? Investing in corporate bonds? This independent guide fro the Australian Securities and Investents Coission (ASIC) can help you look past the return and assess the risks of corporate bonds. If you re thinking

More information

A Hybrid Grey-Game-MCDM Method for ERP Selecting Based on BSC. M. H. Kamfiroozi, 2 A. BonyadiNaeini

A Hybrid Grey-Game-MCDM Method for ERP Selecting Based on BSC. M. H. Kamfiroozi, 2 A. BonyadiNaeini Int. J. Manag. Bus. Res., 3 (1), 13-20, Winter 2013 IAU A Hybrid Grey-Gae-MCDM Method for ERP Selecting Based on BSC 1 M. H. Kafiroozi, 2 A. BonyadiNaeini 1,2 Departent of Industrial Engineering, Iran

More information

Load balancing over redundant wireless sensor networks based on diffluent

Load balancing over redundant wireless sensor networks based on diffluent Load balancing over redundant wireless sensor networks based on diffluent Abstract Xikui Gao Yan ai Yun Ju School of Control and Coputer Engineering North China Electric ower University 02206 China Received

More information

Investing in corporate bonds?

Investing in corporate bonds? Investing in corporate bonds? This independent guide fro the Australian Securities and Investents Coission (ASIC) can help you look past the return and assess the risks of corporate bonds. If you re thinking

More information

A Study on the Chain Restaurants Dynamic Negotiation Games of the Optimization of Joint Procurement of Food Materials

A Study on the Chain Restaurants Dynamic Negotiation Games of the Optimization of Joint Procurement of Food Materials International Journal of Coputer Science & Inforation Technology (IJCSIT) Vol 6, No 1, February 2014 A Study on the Chain estaurants Dynaic Negotiation aes of the Optiization of Joint Procureent of Food

More information

Media Adaptation Framework in Biofeedback System for Stroke Patient Rehabilitation

Media Adaptation Framework in Biofeedback System for Stroke Patient Rehabilitation Media Adaptation Fraework in Biofeedback Syste for Stroke Patient Rehabilitation Yinpeng Chen, Weiwei Xu, Hari Sundara, Thanassis Rikakis, Sheng-Min Liu Arts, Media and Engineering Progra Arizona State

More information

Roland Berger Strategy Consultants. content. Fresh thinking for decision makers

Roland Berger Strategy Consultants. content. Fresh thinking for decision makers Roland Berger Strategy Consultants content Fresh thinking for decision akers Opportunities for obile telephone operators in rural areas Custoers in the countryside are different They need unique offerings

More information

Real Time Target Tracking with Binary Sensor Networks and Parallel Computing

Real Time Target Tracking with Binary Sensor Networks and Parallel Computing Real Tie Target Tracking with Binary Sensor Networks and Parallel Coputing Hong Lin, John Rushing, Sara J. Graves, Steve Tanner, and Evans Criswell Abstract A parallel real tie data fusion and target tracking

More information

Optimal Resource-Constraint Project Scheduling with Overlapping Modes

Optimal Resource-Constraint Project Scheduling with Overlapping Modes Optial Resource-Constraint Proect Scheduling with Overlapping Modes François Berthaut Lucas Grèze Robert Pellerin Nathalie Perrier Adnène Hai February 20 CIRRELT-20-09 Bureaux de Montréal : Bureaux de

More information

REQUIREMENTS FOR A COMPUTER SCIENCE CURRICULUM EMPHASIZING INFORMATION TECHNOLOGY SUBJECT AREA: CURRICULUM ISSUES

REQUIREMENTS FOR A COMPUTER SCIENCE CURRICULUM EMPHASIZING INFORMATION TECHNOLOGY SUBJECT AREA: CURRICULUM ISSUES REQUIREMENTS FOR A COMPUTER SCIENCE CURRICULUM EMPHASIZING INFORMATION TECHNOLOGY SUBJECT AREA: CURRICULUM ISSUES Charles Reynolds Christopher Fox reynolds @cs.ju.edu fox@cs.ju.edu Departent of Coputer

More information

Study on the development of statistical data on the European security technological and industrial base

Study on the development of statistical data on the European security technological and industrial base Study on the developent of statistical data on the European security technological and industrial base Security Sector Survey Analysis: France Client: European Coission DG Migration and Hoe Affairs Brussels,

More information

Example: Suppose that we deposit $1000 in a bank account offering 3% interest, compounded monthly. How will our money grow?

Example: Suppose that we deposit $1000 in a bank account offering 3% interest, compounded monthly. How will our money grow? Finance 111 Finance We have to work with oney every day. While balancing your checkbook or calculating your onthly expenditures on espresso requires only arithetic, when we start saving, planning for retireent,

More information

Machine Learning Applications in Grid Computing

Machine Learning Applications in Grid Computing Machine Learning Applications in Grid Coputing George Cybenko, Guofei Jiang and Daniel Bilar Thayer School of Engineering Dartouth College Hanover, NH 03755, USA gvc@dartouth.edu, guofei.jiang@dartouth.edu

More information

arxiv:0805.1434v1 [math.pr] 9 May 2008

arxiv:0805.1434v1 [math.pr] 9 May 2008 Degree-distribution stability of scale-free networs Zhenting Hou, Xiangxing Kong, Dinghua Shi,2, and Guanrong Chen 3 School of Matheatics, Central South University, Changsha 40083, China 2 Departent of

More information

Exploiting Hardware Heterogeneity within the Same Instance Type of Amazon EC2

Exploiting Hardware Heterogeneity within the Same Instance Type of Amazon EC2 Exploiting Hardware Heterogeneity within the Sae Instance Type of Aazon EC2 Zhonghong Ou, Hao Zhuang, Jukka K. Nurinen, Antti Ylä-Jääski, Pan Hui Aalto University, Finland; Deutsch Teleko Laboratories,

More information

Small Business ebook. 5 Steps to a killer social media strategy

Small Business ebook. 5 Steps to a killer social media strategy Sall Business ebook 5 Steps to a killer social edia strategy About the authors John Keepax and Frank Irias offer ore than 32 years of cobined experience in the areas of John Keepax Creative Director /

More information

Generating Certification Authority Authenticated Public Keys in Ad Hoc Networks

Generating Certification Authority Authenticated Public Keys in Ad Hoc Networks SECURITY AND COMMUNICATION NETWORKS Published online in Wiley InterScience (www.interscience.wiley.co). Generating Certification Authority Authenticated Public Keys in Ad Hoc Networks G. Kounga 1, C. J.

More information

Markov Models and Their Use for Calculations of Important Traffic Parameters of Contact Center

Markov Models and Their Use for Calculations of Important Traffic Parameters of Contact Center Markov Models and Their Use for Calculations of Iportant Traffic Paraeters of Contact Center ERIK CHROMY, JAN DIEZKA, MATEJ KAVACKY Institute of Telecounications Slovak University of Technology Bratislava

More information

Research on Risk Assessment of PFI Projects Based on Grid-fuzzy Borda Number

Research on Risk Assessment of PFI Projects Based on Grid-fuzzy Borda Number Researc on Risk Assessent of PFI Projects Based on Grid-fuzzy Borda Nuber LI Hailing 1, SHI Bensan 2 1. Scool of Arcitecture and Civil Engineering, Xiua University, Cina, 610039 2. Scool of Econoics and

More information

Option B: Credit Card Processing

Option B: Credit Card Processing Attachent B Option B: Credit Card Processing Request for Proposal Nuber 4404 Z1 Bidders are required coplete all fors provided in this attachent if bidding on Option B: Credit Card Processing. Note: If

More information

Data Set Generation for Rectangular Placement Problems

Data Set Generation for Rectangular Placement Problems Data Set Generation for Rectangular Placeent Probles Christine L. Valenzuela (Muford) Pearl Y. Wang School of Coputer Science & Inforatics Departent of Coputer Science MS 4A5 Cardiff University George

More information

Over the last few years, many organizations

Over the last few years, many organizations Evaluating and Selecting Mining Tools by Mike Ferguson Over the last few years, any organizations have invested oney developing data warehouses and data arts. They have ipleented systeatic schees for extracting

More information

Managing Complex Network Operation with Predictive Analytics

Managing Complex Network Operation with Predictive Analytics Managing Coplex Network Operation with Predictive Analytics Zhenyu Huang, Pak Chung Wong, Patrick Mackey, Yousu Chen, Jian Ma, Kevin Schneider, and Frank L. Greitzer Pacific Northwest National Laboratory

More information

Dynamic Placement for Clustered Web Applications

Dynamic Placement for Clustered Web Applications Dynaic laceent for Clustered Web Applications A. Karve, T. Kibrel, G. acifici, M. Spreitzer, M. Steinder, M. Sviridenko, and A. Tantawi IBM T.J. Watson Research Center {karve,kibrel,giovanni,spreitz,steinder,sviri,tantawi}@us.ib.co

More information

Resource Allocation in Wireless Networks with Multiple Relays

Resource Allocation in Wireless Networks with Multiple Relays Resource Allocation in Wireless Networks with Multiple Relays Kağan Bakanoğlu, Stefano Toasin, Elza Erkip Departent of Electrical and Coputer Engineering, Polytechnic Institute of NYU, Brooklyn, NY, 0

More information

Local Area Network Management

Local Area Network Management Technology Guidelines for School Coputer-based Technologies Local Area Network Manageent Local Area Network Manageent Introduction This docuent discusses the tasks associated with anageent of Local Area

More information

Evaluating the Effectiveness of Task Overlapping as a Risk Response Strategy in Engineering Projects

Evaluating the Effectiveness of Task Overlapping as a Risk Response Strategy in Engineering Projects Evaluating the Effectiveness of Task Overlapping as a Risk Response Strategy in Engineering Projects Lucas Grèze Robert Pellerin Nathalie Perrier Patrice Leclaire February 2011 CIRRELT-2011-11 Bureaux

More information

The individual neurons are complicated. They have a myriad of parts, subsystems and control mechanisms. They convey information via a host of

The individual neurons are complicated. They have a myriad of parts, subsystems and control mechanisms. They convey information via a host of CHAPTER 4 ARTIFICIAL NEURAL NETWORKS 4. INTRODUCTION Artificial Neural Networks (ANNs) are relatively crude electronic odels based on the neural structure of the brain. The brain learns fro experience.

More information

Image restoration for a rectangular poor-pixels detector

Image restoration for a rectangular poor-pixels detector Iage restoration for a rectangular poor-pixels detector Pengcheng Wen 1, Xiangjun Wang 1, Hong Wei 2 1 State Key Laboratory of Precision Measuring Technology and Instruents, Tianjin University, China 2

More information

Performance Evaluation of Machine Learning Techniques using Software Cost Drivers

Performance Evaluation of Machine Learning Techniques using Software Cost Drivers Perforance Evaluation of Machine Learning Techniques using Software Cost Drivers Manas Gaur Departent of Coputer Engineering, Delhi Technological University Delhi, India ABSTRACT There is a treendous rise

More information

- 265 - Part C. Property and Casualty Insurance Companies

- 265 - Part C. Property and Casualty Insurance Companies Part C. Property and Casualty Insurance Copanies This Part discusses proposals to curtail favorable tax rules for property and casualty ("P&C") insurance copanies. The syste of reserves for unpaid losses

More information

Markovian inventory policy with application to the paper industry

Markovian inventory policy with application to the paper industry Coputers and Cheical Engineering 26 (2002) 1399 1413 www.elsevier.co/locate/copcheeng Markovian inventory policy with application to the paper industry K. Karen Yin a, *, Hu Liu a,1, Neil E. Johnson b,2

More information

Searching strategy for multi-target discovery in wireless networks

Searching strategy for multi-target discovery in wireless networks Searching strategy for ulti-target discovery in wireless networks Zhao Cheng, Wendi B. Heinzelan Departent of Electrical and Coputer Engineering University of Rochester Rochester, NY 467 (585) 75-{878,

More information

An improved TF-IDF approach for text classification *

An improved TF-IDF approach for text classification * Zhang et al. / J Zheiang Univ SCI 2005 6A(1:49-55 49 Journal of Zheiang University SCIECE ISS 1009-3095 http://www.zu.edu.cn/zus E-ail: zus@zu.edu.cn An iproved TF-IDF approach for text classification

More information

INTEGRATED ENVIRONMENT FOR STORING AND HANDLING INFORMATION IN TASKS OF INDUCTIVE MODELLING FOR BUSINESS INTELLIGENCE SYSTEMS

INTEGRATED ENVIRONMENT FOR STORING AND HANDLING INFORMATION IN TASKS OF INDUCTIVE MODELLING FOR BUSINESS INTELLIGENCE SYSTEMS Artificial Intelligence Methods and Techniques for Business and Engineering Applications 210 INTEGRATED ENVIRONMENT FOR STORING AND HANDLING INFORMATION IN TASKS OF INDUCTIVE MODELLING FOR BUSINESS INTELLIGENCE

More information

Energy Efficient VM Scheduling for Cloud Data Centers: Exact allocation and migration algorithms

Energy Efficient VM Scheduling for Cloud Data Centers: Exact allocation and migration algorithms Energy Efficient VM Scheduling for Cloud Data Centers: Exact allocation and igration algoriths Chaia Ghribi, Makhlouf Hadji and Djaal Zeghlache Institut Mines-Téléco, Téléco SudParis UMR CNRS 5157 9, Rue

More information

Insurance Spirals and the Lloyd s Market

Insurance Spirals and the Lloyd s Market Insurance Spirals and the Lloyd s Market Andrew Bain University of Glasgow Abstract This paper presents a odel of reinsurance arket spirals, and applies it to the situation that existed in the Lloyd s

More information

ADJUSTING FOR QUALITY CHANGE

ADJUSTING FOR QUALITY CHANGE ADJUSTING FOR QUALITY CHANGE 7 Introduction 7.1 The easureent of changes in the level of consuer prices is coplicated by the appearance and disappearance of new and old goods and services, as well as changes

More information

Design of Model Reference Self Tuning Mechanism for PID like Fuzzy Controller

Design of Model Reference Self Tuning Mechanism for PID like Fuzzy Controller Research Article International Journal of Current Engineering and Technology EISSN 77 46, PISSN 347 56 4 INPRESSCO, All Rights Reserved Available at http://inpressco.co/category/ijcet Design of Model Reference

More information

Quality evaluation of the model-based forecasts of implied volatility index

Quality evaluation of the model-based forecasts of implied volatility index Quality evaluation of the odel-based forecasts of iplied volatility index Katarzyna Łęczycka 1 Abstract Influence of volatility on financial arket forecasts is very high. It appears as a specific factor

More information

Factored Models for Probabilistic Modal Logic

Factored Models for Probabilistic Modal Logic Proceedings of the Twenty-Third AAAI Conference on Artificial Intelligence (2008 Factored Models for Probabilistic Modal Logic Afsaneh Shirazi and Eyal Air Coputer Science Departent, University of Illinois

More information

Standards and Protocols for the Collection and Dissemination of Graduating Student Initial Career Outcomes Information For Undergraduates

Standards and Protocols for the Collection and Dissemination of Graduating Student Initial Career Outcomes Information For Undergraduates National Association of Colleges and Eployers Standards and Protocols for the Collection and Disseination of Graduating Student Initial Career Outcoes Inforation For Undergraduates Developed by the NACE

More information

Study on the development of statistical data on the European security technological and industrial base

Study on the development of statistical data on the European security technological and industrial base Study on the developent of statistical data on the European security technological and industrial base Security Sector Survey Analysis: Poland Client: European Coission DG Migration and Hoe Affairs Brussels,

More information

Mobile Backhaul in Heterogeneous Network Deployments: Technology Options and Power Consumption

Mobile Backhaul in Heterogeneous Network Deployments: Technology Options and Power Consumption Mobile Backhaul in Heterogeneous Network Deployents: Technology Options and Power Consuption Paolo Monti*, Meber, IEEE, Sibel Tobaz, Meber, IEEE, Lena Wosinska, Senior Meber, IEEE, Jens Zander, Meber,

More information

A framework for performance monitoring, load balancing, adaptive timeouts and quality of service in digital libraries

A framework for performance monitoring, load balancing, adaptive timeouts and quality of service in digital libraries Int J Digit Libr (2000) 3: 9 35 INTERNATIONAL JOURNAL ON Digital Libraries Springer-Verlag 2000 A fraework for perforance onitoring, load balancing, adaptive tieouts and quality of service in digital libraries

More information

Red Hat Enterprise Linux: Creating a Scalable Open Source Storage Infrastructure

Red Hat Enterprise Linux: Creating a Scalable Open Source Storage Infrastructure Red Hat Enterprise Linux: Creating a Scalable Open Source Storage Infrastructure By Alan Radding and Nick Carr Abstract This paper discusses the issues related to storage design and anageent when an IT

More information

The Application of Bandwidth Optimization Technique in SLA Negotiation Process

The Application of Bandwidth Optimization Technique in SLA Negotiation Process The Application of Bandwidth Optiization Technique in SLA egotiation Process Srecko Krile University of Dubrovnik Departent of Electrical Engineering and Coputing Cira Carica 4, 20000 Dubrovnik, Croatia

More information

Energy Proportionality for Disk Storage Using Replication

Energy Proportionality for Disk Storage Using Replication Energy Proportionality for Disk Storage Using Replication Jinoh Ki and Doron Rote Lawrence Berkeley National Laboratory University of California, Berkeley, CA 94720 {jinohki,d rote}@lbl.gov Abstract Energy

More information

Cooperative Caching for Adaptive Bit Rate Streaming in Content Delivery Networks

Cooperative Caching for Adaptive Bit Rate Streaming in Content Delivery Networks Cooperative Caching for Adaptive Bit Rate Streaing in Content Delivery Networs Phuong Luu Vo Departent of Coputer Science and Engineering, International University - VNUHCM, Vietna vtlphuong@hciu.edu.vn

More information

An Application Research on the Workflow-based Large-scale Hospital Information System Integration

An Application Research on the Workflow-based Large-scale Hospital Information System Integration 106 JOURNAL OF COMPUTERS, VOL. 6, NO. 1, JANUARY 2011 An Application Research on the Workflow-based Large-scale Hospital Inforation Syste Integration Yang Guojun School of Coputer, Neijiang Noral University,

More information

Driving Behavior Analysis Based on Vehicle OBD Information and AdaBoost Algorithms

Driving Behavior Analysis Based on Vehicle OBD Information and AdaBoost Algorithms , March 18-20, 2015, Hong Kong Driving Behavior Analysis Based on Vehicle OBD Inforation and AdaBoost Algoriths Shi-Huang Chen, Jeng-Shyang Pan, and Kaixuan Lu Abstract This paper proposes a novel driving

More information

Implementation of Active Queue Management in a Combined Input and Output Queued Switch

Implementation of Active Queue Management in a Combined Input and Output Queued Switch pleentation of Active Queue Manageent in a obined nput and Output Queued Switch Bartek Wydrowski and Moshe Zukeran AR Special Research entre for Ultra-Broadband nforation Networks, EEE Departent, The University

More information

Is Pay-as-You-Drive Insurance a Better Way to Reduce Gasoline than Gasoline Taxes?

Is Pay-as-You-Drive Insurance a Better Way to Reduce Gasoline than Gasoline Taxes? Is Pay-as-You-Drive Insurance a Better Way to Reduce Gasoline than Gasoline Taxes? By Ian W.H. Parry Despite concerns about US dependence on a volatile world oil arket, greenhouse gases fro fuel cobustion,

More information

ESTIMATING LIQUIDITY PREMIA IN THE SPANISH GOVERNMENT SECURITIES MARKET

ESTIMATING LIQUIDITY PREMIA IN THE SPANISH GOVERNMENT SECURITIES MARKET ESTIMATING LIQUIDITY PREMIA IN THE SPANISH GOVERNMENT SECURITIES MARKET Francisco Alonso, Roberto Blanco, Ana del Río and Alicia Sanchis Banco de España Banco de España Servicio de Estudios Docuento de

More information

Efficient Key Management for Secure Group Communications with Bursty Behavior

Efficient Key Management for Secure Group Communications with Bursty Behavior Efficient Key Manageent for Secure Group Counications with Bursty Behavior Xukai Zou, Byrav Raaurthy Departent of Coputer Science and Engineering University of Nebraska-Lincoln Lincoln, NE68588, USA Eail:

More information

New for 2016! Get Licensed

New for 2016! Get Licensed Financial Manageent 2016 HS There s only one place you need to go for all your professional developent needs. The Power to Know. NEW Experience a different school of learning! New for 2016! Online courses

More information

Preference-based Search and Multi-criteria Optimization

Preference-based Search and Multi-criteria Optimization Fro: AAAI-02 Proceedings. Copyright 2002, AAAI (www.aaai.org). All rights reserved. Preference-based Search and Multi-criteria Optiization Ulrich Junker ILOG 1681, route des Dolines F-06560 Valbonne ujunker@ilog.fr

More information

SAMPLING METHODS LEARNING OBJECTIVES

SAMPLING METHODS LEARNING OBJECTIVES 6 SAMPLING METHODS 6 Using Statistics 6-6 2 Nonprobability Sapling and Bias 6-6 Stratified Rando Sapling 6-2 6 4 Cluster Sapling 6-4 6 5 Systeatic Sapling 6-9 6 6 Nonresponse 6-2 6 7 Suary and Review of

More information

AUC Optimization vs. Error Rate Minimization

AUC Optimization vs. Error Rate Minimization AUC Optiization vs. Error Rate Miniization Corinna Cortes and Mehryar Mohri AT&T Labs Research 180 Park Avenue, Florha Park, NJ 0793, USA {corinna, ohri}@research.att.co Abstract The area under an ROC

More information

Project Evaluation Roadmap. Capital Budgeting Process. Capital Expenditure. Major Cash Flow Components. Cash Flows... COMM2501 Financial Management

Project Evaluation Roadmap. Capital Budgeting Process. Capital Expenditure. Major Cash Flow Components. Cash Flows... COMM2501 Financial Management COMM501 Financial Manageent Project Evaluation 1 (Capital Budgeting) Project Evaluation Roadap COMM501 Financial Manageent Week 7 Week 7 Project dependencies Net present value ethod Relevant cash flows

More information

AutoHelp. An 'Intelligent' Case-Based Help Desk Providing. Web-Based Support for EOSDIS Customers. A Concept and Proof-of-Concept Implementation

AutoHelp. An 'Intelligent' Case-Based Help Desk Providing. Web-Based Support for EOSDIS Customers. A Concept and Proof-of-Concept Implementation //j yd xd/_ ' Year One Report ":,/_i',:?,2... i" _.,.j- _,._".;-/._. ","/ AutoHelp An 'Intelligent' Case-Based Help Desk Providing Web-Based Support for EOSDIS Custoers A Concept and Proof-of-Concept Ipleentation

More information

A Scalable Application Placement Controller for Enterprise Data Centers

A Scalable Application Placement Controller for Enterprise Data Centers W WWW 7 / Track: Perforance and Scalability A Scalable Application Placeent Controller for Enterprise Data Centers Chunqiang Tang, Malgorzata Steinder, Michael Spreitzer, and Giovanni Pacifici IBM T.J.

More information

A decision model for evaluating third-party logistics providers using fuzzy analytic hierarchy process

A decision model for evaluating third-party logistics providers using fuzzy analytic hierarchy process African Journal of Business Manageent Vol. 4(3), pp. 339-349, March 00 Available online at http://www.acadeicjournals.org/ajbm ISSN 993-833 00 Acadeic Journals Full Length Research Paper A decision odel

More information

ADVANTAGES AND DISADVANTAGES OF USING FOREIGN MANAGERS IN IRAN OIL INTERNATIONAL PROJECTS

ADVANTAGES AND DISADVANTAGES OF USING FOREIGN MANAGERS IN IRAN OIL INTERNATIONAL PROJECTS Volue: 02 Issue: 05 Aug-205 www.irjet.net p-issn: 2395-0072 ADVANTAGES AND DISADVANTAGES OF USING FOREIGN MANAGERS IN IRAN OIL INTERNATIONAL PROJECTS Mahoud Gholipoor, Safoora Rashidinejad(Corresponding

More information

The United States was in the midst of a

The United States was in the midst of a A Prier on the Mortgage Market and Mortgage Finance Daniel J. McDonald and Daniel L. Thornton This article is a prier on ortgage finance. It discusses the basics of the ortgage arket and ortgage finance.

More information

Don t Run With Your Retirement Money

Don t Run With Your Retirement Money Don t Run With Your Retireent Money Understanding Your Resources and How Best to Use The A joint project of The Actuarial Foundation and WISER, the Woen s Institute for a Secure Retireent WISER THE WOMEN

More information

High Performance Chinese/English Mixed OCR with Character Level Language Identification

High Performance Chinese/English Mixed OCR with Character Level Language Identification 2009 0th International Conference on Docuent Analysis and Recognition High Perforance Chinese/English Mixed OCR with Character Level Language Identification Kai Wang Institute of Machine Intelligence,

More information

Algorithmica 2001 Springer-Verlag New York Inc.

Algorithmica 2001 Springer-Verlag New York Inc. Algorithica 2001) 30: 101 139 DOI: 101007/s00453-001-0003-0 Algorithica 2001 Springer-Verlag New York Inc Optial Search and One-Way Trading Online Algoriths R El-Yaniv, 1 A Fiat, 2 R M Karp, 3 and G Turpin

More information

A CHAOS MODEL OF SUBHARMONIC OSCILLATIONS IN CURRENT MODE PWM BOOST CONVERTERS

A CHAOS MODEL OF SUBHARMONIC OSCILLATIONS IN CURRENT MODE PWM BOOST CONVERTERS A CHAOS MODEL OF SUBHARMONIC OSCILLATIONS IN CURRENT MODE PWM BOOST CONVERTERS Isaac Zafrany and Sa BenYaakov Departent of Electrical and Coputer Engineering BenGurion University of the Negev P. O. Box

More information

Modeling Parallel Applications Performance on Heterogeneous Systems

Modeling Parallel Applications Performance on Heterogeneous Systems Modeling Parallel Applications Perforance on Heterogeneous Systes Jaeela Al-Jaroodi, Nader Mohaed, Hong Jiang and David Swanson Departent of Coputer Science and Engineering University of Nebraska Lincoln

More information

A short-term, pattern-based model for water-demand forecasting

A short-term, pattern-based model for water-demand forecasting 39 Q IWA Publishing 2007 Journal of Hydroinforatics 09.1 2007 A short-ter, pattern-based odel for water-deand forecasting Stefano Alvisi, Marco Franchini and Alberto Marinelli ABSTRACT The short-ter, deand-forecasting

More information

Reliability Constrained Packet-sizing for Linear Multi-hop Wireless Networks

Reliability Constrained Packet-sizing for Linear Multi-hop Wireless Networks Reliability Constrained acket-sizing for inear Multi-hop Wireless Networks Ning Wen, and Randall A. Berry Departent of Electrical Engineering and Coputer Science Northwestern University, Evanston, Illinois

More information

( C) CLASS 10. TEMPERATURE AND ATOMS

( C) CLASS 10. TEMPERATURE AND ATOMS CLASS 10. EMPERAURE AND AOMS 10.1. INRODUCION Boyle s understanding of the pressure-volue relationship for gases occurred in the late 1600 s. he relationships between volue and teperature, and between

More information

Are you managing your Service Portfolio?

Are you managing your Service Portfolio? White Paper Are you anaging your Service Portfolio? About this White Paper The CIO is expected to be the prestidigitator 1, anaging priorities like a juggler attepting the next world record. But just how

More information