ISDS 2001 CH3

ISDS 2001 CH3 - NOW wciihc tilt"i 51“ lie among...

Info iconThis preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 6
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: NOW wciihc, tilt? "i 51“. lie ' among customers based on c_1;_e_d.i_t_1isk, usage and other CHAPTER 3 — BUSINESS ANALYTICS AND ' DATA VISUALIZATION CHAPTER OBJECTIVES 1. Describe business analytics (BA) and its impedance to organizations 2. List and briefly describe the major BA methods and tools 3. Describe how online analytical processing (OLAP), data visualization, and multidimensionality can improve decision making 4. Describe advanced analysis methods 5. Describe geographical information systems (618) and their support to decision making 6. Describe real-time BA 7. Describe how business intelligence (BI) supports competitive intelligence 8. Describe automated decision support (ADS) systems and their benefits 9. Explain how the Web relates to BA 10. Describe Web intelligence and Web analytics and their importance to organizations 11. Describe implementation issues related to BA and success factors for BA CHAPTER OVERVIEW This chapter goes along with the previous one to present, between Chapter 2 and 3, the entire data warehousing picture for decision support is presented. Chapter 2 covered how the data are stored for access; this one c0vers what happens when a user accesses them. It discusses a wide range of approaches. Students should finish the chapter with a good appreciation of all. 4. Note that an analytical application is a higher level of sophistication when compared to analysis techniques or tools. It includes activities such as: a. automating the thinking of humans and part of the decision making process of humans (remember ADS) b. complex data analytic techniques such as multivariate regression, data mining, artificial inteiiigence, linear programming. c. Example: Credit Scoring for a loan applicant - Calculate a credit-worthiness score - Automatically accept or deny a loan application Select the loan limit - Select the credit card product along with defined interest rate, payment terms, etc. d. More examples of application of analytics 1) Capital One uses data analysis to differentiate characteristics and then to match customer & 9a characteristics with appropriate product offerings. use for it ndnclafl @Ft‘l‘lififlg- 2) Harrah’s, the gaming firm, uses analytics in its W. 3) E & J Gallo Winery qgegliijatixelgnnahm and predicts the appeal of its Eggs. 4) Between 2002 and 2005, Deere & Company saved more than $1 billion by employing a new WW. 7 What was 30m an m "i WM ' " Co was that? 2 I. OPENING VIGNETTE: LEXMARK lNTERNATIONAL IMPROVES OPERATIONS WITH B1 A. Problem B. Solution C. Results D. What we can learn from this vignette For Homework: Answer Section 3.1 Review Questions 1-4 on page 85. ll. THE BUSINESS ANALYTICS (BA) FIELD: AN OVERVIEW A.Introduction — Recall that BI involves gathering data, placing that data in a DW, and using that data for decision making. Business Analysis (BA) is the component of BI that provides models and analysis procedures. B. The essentials of BA 1. Analytics w- the science of analysis 4 ' ' L id: 4—2 Busmess Analytics (3.43 a broad category of applications and techniques for gathering, storing, analyzing, and providing access to data for decision making purposes. Oilfield: ' BA is also known as analytical processing, Bl tools, BI applications, or just Bl. For Homework: Read Application Case 3.1: Ben & Jerry’s Exccls with BA, p. 8'7 — Recall the case (it discussion from Chapter 1 on Ben & Jerry’s. C. The tools and techniques of BA - There are a large number of analytical tools and techniques. 1. These can be divided into THREE major categories See Figure 3.1, page 88. - m a. Information and knowledge discovery: tools that help people take advantage of information in a data warehouse; tools that aid in finding useful patterns/information in data b. Decision support and intelligent systems: the broader application that takes advantage of the analyses performed by tools in the first category. c. Visualization: tools that present the results of the other categories in a way that helps managers understand them and thereby make informed decisions. 2. MicroStrategy classifies BA tools into FIV n .- . 'Pivej‘\ l a [pm 110‘ K may (finish-ms a has. managers DES wk (1’ 10' W wradWS. gal-rm. Clfl'ifilW-afl‘i rill dz 2 5 a new d. statistical anrgyjrs and data mining insets 2m rt Fina.an lam 5| 4 'e. rec elv ana 1.—"‘ J! , +9 law Maegr‘tfiu bu. Rigt-rtn‘cfieirfitbg «if 81:: 565mg! Ear. hers Dial events as target on Note: Figure 3.2 Illustrates Microstra egy s integrate BI solution, Microstrategy 8, constructed so that all tools are available in one place. new: Students: You may omit SAP’s classification, p. 89. D. Executive information and support systems 2. Online Transaction Processing (OLTP) is concerned with simple, repetitive activities that update small parts of a databme and are often performed hundreds of times per minute or even more irequently. 3. As a result, the best database organization for one may not be good for the other and many organizations use different databases to support the two. C. The major characteristics of OLAP: l) Examining many data items 2) Dealing with complex relationships 3) Looking for patterns, trends, exceptions 4) Allowing users much flexibility in defining their questions a}? D. OLAPs major benefit - helping users to find what they want to know by using a database (or DW). E.Types of OLAP — The major types of OLAP are as follows. «it. engaged-€399 categories: . t ‘ n‘ —; l hr after} stain-«3r are a as as? retreat“ at” b- Cilbe anal Sis - OLA? mulir‘dcm . Sire: cites, analta’owfi it» “(Edit . ' mo 9mg I‘Et r in A {Mitten-i mam-h 7x 9 signage“ ctr-r fer armed 1. E18 (Executive Information System) is a decision support system that focuses on top executives. It serves their information needs by providing a user-friendly interface through which executives can access timely and relevant information, drill down into it, and seems management reports. 2. E85 (Executive Support System) adds capabilities such as intelligence and analysis support, communications, and office automation to an EIS to support the complete information processing needs oftop management. nest if“ V- 3. Major capabilities of EIS/ESS (refer to Table 3.1, page 88) are drill- mind status reports, trend analysis, ad hoc analysis, exception reporting, and slicing & dicing. u and anal sis . down, critical success factors (CSF), key performance indicators KPIs , a assassins a 4% unto Parnassus/y < ) lzr l ii“ r Diw For Homework: Answer Section 3.2 Review Questions 1-7 on atqf-ES sagsflJ'Q‘hbnfi page 91. a" S ' "5"“: “PM” Watch this video on your own to reinforce the notes: OLAP Cartoons, Business Intelligence, Dimensions, Measures httpfilwwwyoutube.com/watch?v=KTdJ66101 3s&NR=l III. ONLINE ANALYTICAL PROCESSING (OLAP) A. Introductions — Online analytical processing (OLAP) refers to a variety of methods performed by end users in online systems. These methods include generating queries, requesting ad hoc reports, conducting statistical anaiyses, and building visual representations (like dashboards and scorecards); OLAP provides a multidimensional view of the data. + at -i ‘5 not mxiapanci Wm B OLAP Versus 0er imam“)? mfg “gmwmfgd' " 0 l. OLAP is concerned with conipl x ana 3195 t t read, but not modify, large parts of a databasei apdflgtpafgfln-ea‘fegpflneql M “WIFE 4‘ _ a? A“ _ infigpimently. 41.04% cmtoipmgégfimjw 'tfl‘ry'. , 3‘? our—9 rpvoiiur 's rpm,{(1.9“m3mimmgg, anar‘sgj rerr‘msutrfiearurt - Tier? which to "9+ Drama. A—ll 05 three are Gui-i? exempt Far . . . I N 10$ 7 p -.rwu'+rvu porcine-Is / or; ’we :31..th no new; A», {Quip - Paleocene Ea Waldsarmmwwnfl a A? q [ton’l' - 4. Web OLAP (WOLAP): OLAP in which data are accessed via a Web browser. name. 5. Desktop OLAP: OLAP in which data are stored, and anal sis is erformed,onauser’s eskto s stem. or. lazupnpo, simple. OLA? 4min. p y F. Examples of successfiil OLAP applications (see Table 3.2, page 93) ' G. OLAP tools and vendors For Homework: Answer Section 3.3 Review Questions 14 on page 95. For Homework: Read Application Case 3.2: TCF Financial Corp.: Conducting OLAP, Reporting and Data Mining, p. 95 a IV. REPORTS AND QUERIES (Note that they are not mutually exclusive.) V! 31.11% Adntroduction — OLAP uses reports and queries. The result of a to W“ 1. Multidimensionai OLAP: OLAP in which data are stored query is often a report, and a report may call for a query. in a multidimensional Engages: rather that;l relatiorl’ltajl4 or dz fl“ Reports can be routine or on-demand (ad hoc). There are ,aa’orane‘ : Starch“ .. . or an.tsn.di“““‘“"” am my” mm m mm mm been at am . ' 2. Relational OLAP (ROLAP): OLAP in which data are E. Reports - Accordin to Codd’s rules, OLAP must be a I g stored in a relational database (rather than vflfifififi'flgfigflgfi use uniform, flexible, and adjustable. mu tidirnensional rSome oth 1‘ one . Shots. gamma 7.1. ,- Miami/fl. to J)” ‘t' goingsiairr'u ‘a‘H-r'wi-er '-”° . /X 3 lééeévelgsiui‘l fins-tidal.“ W.‘ éflfilfinwé; gaggin‘ra 1- Two types Ofreports are: ’5’ Efliaig aha r ._.‘- ..._ . waflz' . ..r .531 xkamtr Routine Reports — generated automatically and 1117;?) a. distributed to subscribers on a period basis. Issued regularly without a specific request each time (See Example Page 96 about a store manager’s reports and enterprise report). Reports can be generated directly from an operational database (ex: POS — point-of-sales systems) and/or from a DW. b. Ad Hoc (On-Demand) Reports — created for a specific user when needed. Ad Hoc reports are created for a specific need, oflen a one-time reguegt..These can be 51 ar to routine reports, but for different time intervals (subsets of time intervals) or for only a subset of the data. (Example: provide a list of all customers who purchased a company’s products for more than $5000 each during January) 2. Multilingual Support — reports can be translated to other languages 3. Examples of Vendor Products a. Business Objects’ Crystal Reports b. Microstrategy c. Cognos 8 Business Intelligence (now owned by IBM) d. Hyperion c. Microsoft Report Builder htth/mvwmicrosoi’t.corn/splserver/2ODS/en/us/rem‘tingaspx H Select * From CrashgTB Where Parish_CD = “01" This SQL statement would return all columns from the crash table where in Acadia parish. D. Analysis of reports’ results must be done quickly so appropriate actions can be taken. View video in class: Microsoft SharePoint— a product for online collaboration, file sharing and web publishing. For Homework: Answer Section 3.4 Review Questions 1-6 on page 99. V. MULTlDlMENSlONALITY - A way to organize data efficiently for analysis and presentation. This approach treats each characteristic of the underlying entity as a dimension of a cube. A cube in an OLAP context can generally be two-dimensional, three-dimensional or higher. A. Multidimensional presentation 1) Three factors in multidimensionality a) Dimensions H such as products, salespeople, market segments, business units, geographic locations, (mflfi, “if” ' , .. . distribution channels, countries, and industries. yfl’i/ b) Measures — such as money, sales volume, head at. 3am count, inventory, and actual versus forecaster] profit. c) Time - such as daily, Weekly, monthly, quarterly, and yearly. 2) A muitidimensional database created from the DW and has data organized by their dimensions to support easy and quick multidimensional analysis. 1'!) 4. Representative Types of Reports (Table 3.4, page 97) — Reports cover all facets of business I 4 arafiri arm Ferry?“ 4' 5. Scorecards and Dashboards —— oflEn a better way to view the information. They bring out important issues more clearly. a. Tabular and graphical views b. Comparisons to metrics and KPls (key performance indicators) c. Aids in data visualization $5) Winn Van Him-s Ind-w “MW—aw Gram}: From u-mszerrrgmm, .EMC‘Cot-parrrlim \ C. Reports and Queries it 3““ Warm m 6mm, . Ad Hoc Queries - on t tcardiot be prgiiicted and e r E is: therefore cannot be planned for; can be written in a O 6 ,1 g \\ language such as SQL or by a query tool. Query tools - \O - inciude menus, query-by~exarnple and natural ’3‘? language interfaces. *3 we“? "6‘ 2. Structured Query Language (SQL) for Querying — Galina ' Taught extensively in ISDS 3110 Data Base course. SQL Query Example in its simplest form: 3i: Sin. is used 4 gourd-11% cougar. \\ I2 3) Example of how it works at the top of page 100. B. Multidimensional data cubes and their analysis —« data cubes allow quick generation of queries and reports by end users. See Figures 3.3 and 3.4 on page 101. DA data cube is a structure used to represent data along several measmegpfiingggsghlt can be two-dimensional, three- dirnleiii‘sional, or higher-dimensional in which each dimension of the data represents a measure of interest. 2)According to one definition in the text, cube is a common short form ofdara cube. 3) According to another, a cube is a subset of interrelated data organized to allow users to combine attributes and metrics in any desired way, such as sales, profit, units, or age. 4) Cube analysis lets goggle perform queries by flipping thing? as H! | games ogfleporgygemws, usrng software features "(spy \ as ivot ,tnsssszbsrrzisstasoii'filr gimme/drill- risen. . I our used is (On-mar {GWW‘ m“ " 5) Slice-and-dice data a anal e a cube by using simple mouse 5“ i6 clicks. Rearrange data so that it can he viewed” Eom different . W... m n "Wu-"r l) The term is typically used with OLAP databases that '5 present information to the user in the form of multidimensional cubes similar to a 3D spreadsheet. 2) Crash data — we sliced and diced it using PowerPivot by parish, day of week, and alcohol. 43:91 13 1'4 ,r,//l,. analyze data in more sophisticated ways than simple reports‘and queries. Tools include hypothesis testing, multiple regr_e_ssion and correlation, churn redictions and casting? scoring models. / VI. ADVANCED BUSINESS ANALYTICS - Tools that are able to These are necessary because making good decisions can require more complex analysis than reports and queries can offer. sen-w) A. Data mining - extracts hidden information from databases *4 and finds patterns in damcision maker may not know what he or she is looking for. . \‘h‘t Wi) OLAP is guided by people: data mining by algorithms. Anatvsls Cuties mewf‘,‘ ....... Ami-Ewan“ Muppfiw t g g 2) See Chapters 4 & 6 for more details. Predictive Analysis - help determine the probable future outcome for an event or the likelihood of a situation , , occurring, ct. MWQ‘HCflflg are. Insuring; repays-mg state. we. 1) Aside from prediction: “if interest rates rise by one percent, what will happen to the cost of our raw materials?” these tools can also identify relationships and patterns. 2) Data mining is concerned with finding relationships that existed in the past. Predictive analysis is future-oriented: asking what, based on the data available, will happen if a given fisture event occurs? Sli(e s Diceyiews / B. +PZO €1.05 minim cradrt-mu-n I Pram anaemia- dates C.Limits of dimensionality (such as cost and data storage requirements) are minimal today because of improved technology and the increased business value it generates. For Homework: Answer Section 3.5 Review Questions 1-5 on Page 192- , C. Tools of predictive analysis - clustering, decision trees, market-basket analysis, regression modeling, neural nets, genetic algorithms, text mining, hypothesis testing, decision analytics, and more. "T a ‘ 5‘ HOT Gt gétgrmdnrcdt {0011}: A a .2 at; 3 are I; i- t»? l at! matte. 1'5 [6 D. Predictive analysis is useful in finding activity patterns that predict hand. For example, one might know that a large number of international calls to a particular country on a calling card, absent any prior history of such calls on that card, may indicate possible fi'aud and justify confirming that the calls were legitimate. fiDK/isualization spreadsheets —enables instant views of complex data in a single picture. Spreadsheet visualization is the application of visualization tools to data created in spreadsheets, such as Excel 2010 and PowerPivot; see examples on page 107. EflNew directions in data visualization: E. Representative vendors’ tools for advanced analytics — a list appears on page 105. l) Dashboards and Scorecards — busy executives can gain insight at a glance. For instance, numeric results can be displayed in red, yellow, and green to indicate their status. See Chapter 5. Visit Microstrategy a lications/S St les/scorecards dashboardsas . - . v so ya 2) a comm” hat-Hm 0% visuatimtmnsasmfixgp [Wait Homework: Read Application Case 3.3, page 103, “Predictive Analysis Can Help You Avoid Traffic Jams” http://inrix.com and http://teleatlas.corn For Homework: Answer Section 3.6 Review Questions 1-5 on page 105. Financial data vis alization covers the visualization of many types of financially-related data. a) A CFO may need to view risk, return and liquidity in combination. b) Investment managers can use it to view characteristics of potential investments or of a client’s portfolio. VII. DATA VISUALIZATION ~ Technologies that support visualization and sometimes interpretation of data and information. flan-t dado visunéizatim. "first "list ova-“a that . . - - - n - if?» it Z’ZA' Ipcltgglesaldigmzldmagis’ (:IS’ graphgiai meidlmerfacgs’ aptlis’ ‘ 'f _. W _ F. Vendors and products —New tools are emerging to “ml Fe lty’ menswna magenta 0115’ V1 $03” an amma load“? “32‘? facilitate nontechnical users at the executive level to glean A a ‘ I Q .{4 . . . 9 . . B. Visual tools can help identify relationships such as trends. Q‘sfiffiw :igilggéfifmmatmn from busmess data make demmfls’ alld C. Advantages of data visualization over other methods of presenting data include the ability to identify trends quickly, to For Homework: Read Application Case 3.4, page 109, Financial . . Data Visualization at Merrill Lynch” it? 99 ‘ detect problems that are hard to see otherwise (perhaps havmg . . . 1 gone undetected for years for this mason), and supporting the Answer Section 3.7 Revrew Questions -3 on page 109. 7 use of Web browsers to look at vital data. ,9} Eli, ' ‘ What is di'mitr'inrnvmgfi pheasant-crimes type oi) ant-a. waitresses“- 17 1'8 X<I[[.GEOGRAPHIC INFORMATION SYSTEMS (GIS) — a ii. Uses GIS in its police departments to find b computer-based system for capturing, storing, modeling, geographic patterns in crimes as well as to deploy . retrieving, checking, integrating, manipulating, analyzing, and officers. ‘ 1r) g B displaying geographically referenced data by using digitized iii. Uses GIS in physical asset management. “in, maps. See http://gpsy.com/n1aps/ I J c. GIS projects ongoing for Louisiana Highway Safety JfiThe most distinguishing characteristic of GIS is that every record Research Group. These aid policy makers and Louisiana "‘ ’ or digital object has an identified geographic location. State Police. For more details, see Dr. Hehnut Schneider and Dr. Omer Soysal in the LSU ISDS department. Visit Baton Rouge SQL Server Users Groupss regent speaker httpuzilwmvbus.lsu.edu/academics/isds/facultywcbpagea about spatial data _h_ttp:/fbatonrouge.sglpassorgf‘ M93 Toplc: Building Spatial Data Reports with SQL Server 2008 R2 Reportan Servlms Overview: Miami} introduced spatial data types in SQL Sewer 2008 and since has enhanced the reporting capabilities in the following release of SQL Server 2003 R2. This session will introduce the concept of spatial data type in 5121, provide some tips and tools for getting spatial data Into SQL, and show different ntettrods for includan spatial data In SQL Reportier d. Facebook application: Places Seances stm- B. GIS can support decision making in many applications that Speaker: Hark Verret relate to locations. 310: Mark Venet Is the systen-s Mmlnlstrarar fur the LSU Highway Safety Raearch Group. He has over 10 years ofemertence in the tTfletd. He hm presented armepast two SQLSaburday arena to Baton Rouge. He 618 can be combmed (Global Positionng has functioned as can for the past 3 years working with SQL zone. zoos. zoos, and zone to. Mark Verret and Cory Hubchtmon will! be teaching tsos 3070 steam intelligence, Data Mining Tools, Analytics a para Systems) to create many applications, especially those related to transportation. warehouslng in Spring 2011. A prerequisite is ISDS 3110 Database Prooas Management (/1 mflghfgifggfgi GIS is the ability to integrate geographic D. Usefirl Applications: 018 use location information. GPS up I informatiOn with other types of information. This, in puivides this mfonmifion for a suitabiy equipped inflame NEE? M. - turn, enables decisions that are based in part on Objbia't“EagflatggéeE—iiguzk’fifir example—Es the 31$)" @fi' geographic (location-based) data and in part on other su Jec 0 a ’ pn 1' e ' \ data' E. LSU Tiger Traits — Transit Visualization «- real-time . . display of buses along routes and stops. / b. State of Louisnma , (,3 i. Using WebFocus from Information Builders to http'msu'mmc'com/ " identify individuals who traffic in food stamps. w.) Lafitrsl‘mrta it wot, (Sagesz . _, / Fact—boot. 5 ’Pl Mg}: . Mil/l "l? tL-i‘i'éli ‘1’ ' w {9 20 F. GIS and the Intemet/intranets — Amazing GIS/GPS applications are being created for the Web and Mobile Applications. For Homework: Read Application Case 3.5, page 112, “GIS and GPS Track Where You Are and Help You with What You Do” Answer Section 3.8 Review Questions 1-5 on page 113. TM gmtsms i’édmologltzs {has ewe Qfloséig mawa 1,- ext. IX. REAL-TIME BUSINESS INTELLIGENCE, AUTOMATED DECISION SUPPORT (ADS), AND COMPETITIVE INTELLIGENCE ‘A. “Iggy-time Bi 1) Uses current, rather than historical, data 2) Many business decisions need to be done in real-time, or sti%D' Two concerns about real-time BI both relate to data “a generated hours, or even minutes, later. 2) A related one is that some data may not need continuous updating, so money may be wasted on unnecessary updates. Automated degisjgp filament (ADS) uses We; which are generally executed with intelligent systems, to support large numbers of repetitive decisions, such as price optimization and product configmtign. Bengitts of ADfi', l) Rapidly builds rules-based applications to automate or guide decision makers, and deploys the applications into 1) One is that one person's results may not match another’s $ «0 pg. almost any operating environment. at? 2) Injects predictive analytics into rules-based applications, increasing their potency and value. very close to it. 3) Provides decisioning services to legacy systems, 3) Supporting such decisions with BA requires a real-time expanding their capabilities white minimizing technical DW and special BA- features. .. i , risk. fl?) ‘ B. Bgrefim of Real-time BI , Oi“ A 299’ 4) Combines business rules, predictive models, and 'J- 3‘ i I) It enables people to use BI tools for decisions that ex , man [131". optimization strategies flexibly into State-of-the-art require up-to-date information. +0 (Maid decision-management applications. 2) This extends the usefirlness of BI technology to many 4: 4 . .5» 5) Accelerates the uptake oflcarning from decision criteria more decisions than it was previously able to support Oi mi: N" A into strategy design execution, and refinement 3) The benefit is making this new class of decisions better G. Yield management is the process of adjusting the price of a than they could be made without BI. perishable product, which has limited availability, based on . C. Technologies needed for Real—time BI demand. Its most commonly cited application is in pricing ,i llThe key technology is real-time data warehousing, often ’ airline seats. I with the assistance of automated data acquisition. fill Major Categories of ADS listed in the text include: ADS mp, Ms k 2) They permit the use of data warehousing and BI tools for I)Product or service configuration — Customize a ’PC, not mp Mutt- , real-time operational decisions. 2)Yie1d (price) optimization _ Wm “My 4. Ii 5 “Mt it rmO-l—r'mra 'Z‘Dw is a mgmflg ii a hm' “PM rans- 21' 3) Routing or segmentation decisions 'h M "law-isle Ins. alumni}? 3w» mcrtgaegs, Aendlé the an 4) Corporate and regulatory compliance ‘prc 5) Fraud detection — 125, US SEQ , bani-:ng Lind,“ “u loan: a% 7) Operational control — I. BI tools were designed to support business nalysts in decision making, while ADS systems are designed to make decisions without human intervention. I. However, the decisions made by ADS tools are not those that a business analyst would otherwise make. They are lower-level, operational decisions. ,g, 32 Competitive intelligence is infonnatipn acquired‘about com etitors 1n the come of momtormg their aCtIVitleS. p a . Competitive Intelligence can drive busmess performance g, as . by 1) increasing market knowledge 2) improving knowledge management 3) raising the quality of strategic planning. M. The Internet is an outstanding source of competitive information. Aside fi'om information deliberately made public by competitors, the Internet can be used to find: i ' 1)information inadvertently made public 2) information made public by customers and suppliers 3) information required to be made public by regulatory and other government agencies 4) information posted by those who are upset with the company or have some other axe to grind. For Homework: Answer Section 3.9 Review Questions 1-9 on page 119. x. BUSINESS ANALYTICS AND THE WEB: WEB INTELLIGENCE AND WEB AN ALYTIC S 23 For Homework: Answer Section 3.16 Review Questions 1-5 on page 122. Read Application Case 3.6, page 121, “Web Analytics Improves Performance for Onliue Merchants” httpfl/wwwurcwport— news.c0m/dcfault.aspx XI.USAGE, BENEFITS, AND SUCCESS OF BUSINESS ANALYTICS A. Usage of BA — Almost all medium to large companies use some type of BA, with or without a data warehouse, for significant cost savings. B. Benefit of BI and BA is the ability to analyze data in a data warehouse or elsewhere to support decision making. C. Success and usability of BA — To ure success of BI and BA % - projects, it is important to i_c_l_e_n.ti success factors e.g., user participation) and cultivate those factors. D. i hge key factors that affect the implementation of B1 are puking, pegple and processes. (These are important in the implementation of nearly any other kind of information system, too.) was B. Some BA and BI Best Practices: 1) Focus on Business Value 2) Dashboards & data must be actionable to have concrete value 3) Inclusion of BI-experienced Architect early on — This guarantees proper cube design, which details many projects 22 Manda? 1 all? \- \ can, rust. or. p; 357 Using the Web in BA (DA, | 3 “m ’3 5"” “tulmi'ié’ fi)4mmmnflA_%fiphammflgfi%ged 6) Dynamic forecastingmenui'qgtmps align produriiun ml their Custom; 9m costslfmv.) WTW - the Wabpfvl’l 8 am annvm, W MWW‘M» teatime-tum a‘ciifi’l’m com; swig ,- parmqs Q0 Wm 1.9%; :2)BI activitiesmfrom data acquisition, through warehousing, BA, and mining-hare mostly performed using Web tools. 3) These tools enable users with browsers to log on to a system, make inquiries, get reports, and so on, in real time. B. Web analytics, or Web Intelligence 1) Refers to analysis of Web data (known as clickstream data). 2) Such analyses are useful in market research and competitive intelligence. 3) Users’ behavior patterns at Web sites are important information for those who would like to design Web sites that motivate certain types of behavior. Clickstream analysis refers to the analysis of data that occur inside the Web environment. 1) These data, known as clickstream data, provide a trail of the user’s activities and show the user’s browsing patterns: a) which sites are visited b) which pages are accessed c) how long is spent at the site, and so on. 2) How are clickstream data used? By analyzing and interpreting clicksb‘eam data a furn can, for example, find the effectiveness of promotions and determine which products and ads attract the most attention. Note: To leam more visit: http:#wwwwebanalyticsassociationorgz 24 F. Why BI/BA Projects Can Fail - The text offers this list often: 1. Failure to recognize BI projects as cross-organizational business initiatives and to understand that, as such, they differ from typical standalone solutions 2. Unengaged or weak business sponsors 3. Unavailable or unwilling business representatives fiom the fimctional areas ‘ 4. Lack of skilled (or available) staff, or suboptimal staff utilization 5. No software release concept (i.e., no iterative development method) 6. No work breakdown structure (i.e., no methodology) 1. No business analysis or standardization activities 8. No appreciation of the negative impact of “dirty data” on business profitability 9. No understanding of the necessity for and the use of metadata 10. Too much reliance on disparate methods and tools Read Application Case 3.7, page 124, “Retailers Make Steady BI Progress" For Homework: Answer Section 3.11 Review Questions 1-4 on page 125. ...
View Full Document

{[ snackBarMessage ]}

Page1 / 6

ISDS 2001 CH3 - NOW wciihc tilt"i 51“ lie among...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online