(0 Item)

Business Intelligence and Data Warehousing (BIDW 2010)

Academic Conferences - Business Intelligence and Data Warehousing (BIDW 2010)
Academic Conferences

By : Global Science & Technology Forum

Date : 2010

Location : Singapore / Singapore

PDF 144p
Description :

Latest academic research and business applications on business intelligence, data warehousing, data mining, statistical analysis, forecasting to turn data from its raw form into something usable that business decisions can be based on.

Keywords :

data warehousing, data mining, decision support systems, Consumer Modelling, High-Frequency Data, Conscious Knowledge, Subconscious Knowledge, Business Intelligence, data warehouse, model-based diagnosis, Demographic and behavioural profiling, Data mining methodology, Cloud BI, business intelligence conference proceedings

Keywords inside documents :

business ,intelligence ,system ,model ,knowledge ,models ,mining ,missing ,figure ,imputation ,integration ,charts ,users ,cloud ,financial ,demand ,process ,ratio ,decision ,level

Proceedings from Business Intelligence and Data Warehousing (BIDW 2010) conference.

 

Annual International Academic Conference on Business Intelligence and Data Warehousing (BIDW) is a catch phrase that was coined in the mid-1990s to describe taking data from its raw form and turning it into something usable that business decisions can be based on. It is an umbrella term that ties together other closely related data disciplines including data warehousing, data mining, statistical analysis, forecasting, and decision support.

 

 

 

 

Abstracts of Proceedings included in the Business Intelligence and Data Warehousing (BIDW 2010) conference proceedings. 
 
  1. A Pattern-Supported Approach for Conceptual Modeling of Dimensional Data Models

Narasimha Bolloju, City University of Hong Kong, Tat Chee Avenue, Kowloon, HONG KONG & Yulin Fang, City University of Hong Kong, Tat Chee Avenue, Kowloon, HONG KONG

 

Dimensional data modeling is an important activity in the development of business intelligence systems, especially when a data warehouse or data marts are to be created. Dimensional data

modeling involves identification of required fact and dimension entities that support various types of BI functionality. Since the process of developing dimensional data models differs significantly from that of developing data models for online transaction processing systems, practitioners and researchers have been suggesting approaches to aid this process. Commonly employed approaches for dimensional data modeling make use of operational data models or enterprise data models in identifying grains, dimensions and facts. This paper presents an approach for conceptual modeling of dimensional model with the analysis patterns especially for the analysis of source data models in identifying facts and dimensions, and for the depiction of designed dimensional models. The proposed approach is illustrated for analyzing source data models of an airline company and for developing dimensional data models. The application of analysis patterns is expected to result in an efficient and effective modeling process by facilitating communication among the stakeholders.

 

 

  1. Estimating Consumer Demand From High-Frequency Data

John Cartlidge∗, University of Central Lancashire, Preston, PR1-2HE, United Kingdom & Steve Phelps University of Essex, Colchester, CO4 3SQ, United Kingdom

 

Price discrimination offers sellers the possibility of increasing revenue by capturing a market’s consumer surplus: arising from the low price elasticity segment of customers that would have been prepared to pay more than the current market price. First degree price discrimination requires a seller to know the maximum (reserve) price that each consumer is willing to pay. Discouragingly, this information is often unavailable; making the theoretical ideal a practical impossibility.

Electronic commerce offers a solution; with loyalty cards, transaction statements and online accounts all providing channels of customer monitoring. The vast amount of data generated–eBay alone produces terabytes daily–creates an invaluable repository of information that, if used intelligently, enables consumer behaviour to be modelled and predicted. Thus, once behavioural models are calibrated, price discrimination can be tailored to the level of the individual. Here, we introduce a statistical method designed to model the behaviour of bidders on eBay to estimate demand functions for individual item classes. Using eBay’s temporal bidding data–arrival times, price, user id–the model generates estimates of individual reserve prices for market participants; including hidden, or censored, demand not directly contained within the underlying data. Market demand is then estimated, enabling eBay power sellers–large professional bulk-sellers–to optimize sales and increase revenue. Proprietary software automates this process: analyzing data; modelling behaviour; estimating demand; and generating sales strategy.

This work is a tentative first step of a wider, ongoing, research program to discover a practical methodology for automatically calibrating models of consumers from large-scale high-frequency data. Multi-agent systems and artificial intelligence offer principled approaches to the modelling of complex interactions between multiple individuals. The goal is to dynamically model market interactions using realistic models of individual consumers. Such models offer greater flexibility and insight than static, historical, data analysis alone.

 

  1. Conceptualizing a True Business Intelligence System

Anjan Kumar Swain, Indian Institute of Management Kozhikode, IIMK Campus P.O., & Shyam, A.V. Indian Institute of Management Kozhikode, IIMK Campus P.O.

 

This paper concentrates on the development of a framework to build a true business intelligence system to represent natural intelligence of decision makers in any organization. Further, to incorporate intelligence in a system, a detailed discussion on various factors associated with intelligence are discussed. Also, this paper presents an analysis of the interrelationship among data, information, knowledge, and intelligence. In addition, a classification of knowledge is presented to understand the intelligence building process. Finally, a framework for building a true business intelligence system is provided in line with the development of the adaptive business intelligence system.

 

 

  1. Intranet: How is Business Intelligence Stored in a Data Warehouse?

Lisa Soon, School of Information & Communication Technology, Central Queensland University Mackay Campus, PO Box 5606 Mackay MC, QLD 4740 Australia, & Campbell Fraser, Department of International Business & Asian Studies, Griffith University Logan Campus, Griffith University, University Drive; Meadowbrook QLD 4131 Australia

 

Business intelligence has long been regarded as a critical success factor for contemporary business. While its importance is known, research on what data or information is related to business intelligence and how business intelligence is captured for use in an organizational information system such as a private intranet for use within a wider organizational community is limited. This research discusses how business intelligence is gathered, created, used and realized through the storing in an organizational data warehouse in an intranet. A case study was conducted to examine how an Intranet system was designed in a selected organization, how business intelligence is collected and organized, and how it is used for strategic planning, decision making and daily business operation purposes. This research adopts a qualitative case study method using interviews and observation techniques. The respondents explained how structured business intelligence data was categorized and disseminated to users, and how the used information empowered staff in work performance. It was discovered that the design of the intranet directly relates to retaining knowledge of the staff within the organization, drawing

all internal resources together, capturing resources from external sources, and forming a repository of organizational assets for common use embedded through organizational work procedures within the intranet.

 

  1. Log mining for query recommendation in e-commerce

Gokhan Capan, AnadouUniversity & Alper Kursat Uysal, AnadouUniversity & Ozgur Yilmazel, AnadouUniversity

 

Searching is very important in e-commerce application, especially if its content is generated by users. A query recommendation system, which recommends new queries to a user after his initial query, plays a key role for shortening the search session by guiding the user to reach the product they need. Because the amount of potential queries is nearly infinite, query recommendation can be consifered as an information filtering processn which may either be content-based filtering or collaborative filtering.

In this paper, we desrcribe hybrid method for performing scalable, fast responding query recommendation system, which is combination content-based filtering and collaborative filtering techniques. We show how the method has been applied to a set of user log data collected from an e-commerce site by creating model in two-monthd of query logs and evaluated the model using separate set of querylogs of one-month period. Finally, since (i) most og yhe query seassions are shhort, (ii) the goal of the query recommendation system is keeping the session short, and (iii) traditional evaluation metrics suffer from the lack of rated data; we introduce a new evaluation metric for query recommendation. The new metric returns closer results to humain-based evaluation measures. Results obtained from the evaluation show that the system decrease the search path, and allows users to reach product with fewer cliks.

 

 

  1. Debugging Support for Data Warehouse Applications

Peter Struss, Tech.Univ. of Munich, Boltzmannstr. 3, D-85748 Garching, Germany & Vikas Shivashankar, 3264 A.V. Williams Bldg, Desk #4, University of Maryland, College Park, MD 20742 USA & Mohamed Zahoor, Ericsson India Private Limited, Beliciaa Towers, No.94, Mrc. Nagar, Chennai 600 028, Tamil Ndai, India

 

The paper describes a model-based approach to developing a general tool for localizing faults in applications of data warehouse technology. A model of the application is configured from a library of generic models of standard (types of) modules and exploited by a consistency-based diagnosis algorithm, originally used for diagnosing physical devices. Observing intermediate results can require high efforts or even be impossible, which limits the discriminability between different faults in a sequence of data processing steps. To compensate for this, fault models are used. This becomes a feasible solution for standard modules of a data warehouse application along with a stratification of the data. Fault models capture the potential impact of faults of process steps and data transfer on the data strata as well as on sets of data. Reflecting the nature of the initial symptoms and of the potential checks, these descriptions are stated at a qualitative level. The solution has been validated in customer report generation of a provider of mobile phone services.

 

  1. Identifying the Needs and Benefits of an Electronic Medical Record (EMR) System through Research, Development, and Implementation

Arshia Khan, Department of Computer Information Systems, The College of St. Scholastica & Arif Ahmed, SOWED, India. & Kimona Kousins & Ritika Jhangiani & Shinjini Kar & Michael Lewis, The College of St. Scholastica

 

India is a developing country with an advanced technical base and skill set. Although the technical growth is large, medical provision to the entire country is still a challenge. The low income population in rural and urban areas is often deprived of organized medical facilities, resulting in improper and inadequate patient care and medical development. To resolve this matter we propose the implementation of an EMR system. An EMR system is “an electronic record of health-related information on an individual that can be created, gathered, managed, and consulted by authorized clinicians and staff within one healthcare organization.”(NAHIT, 2008, 6). This system is an efficient method to organize the healthcare system in the rural areas to provide patient care under monitored, documented and advanced conditions.

 

 

  1. Using data mining process standards: A case study - Profiling Internet Banking Users in Jamaica

Gunjan Mansingh, The University of the West Indies, Department of Computing, Kingston, Jamaica & Lila Rao-Graham, The University of the West Indies, Mona School of Business, Kingston, Jamaica & Kweku-Muata Osei-Bryson, Virginia Commonwealth University, School of Business, Richmond, Virginia, U.S.A. & Annette Mills, University of Canterbury, Department of Accounting and Information Systems, Christchurch, New Zealand

 

Internet banking can provide a number of benefits both to the bank and its customers, however, there is a cost associated with providing this service and internet banking will only be

financially viable if these costs can be justified. Justifying these costs requires an understanding of the likely customer base that will take up this banking option so that banks can target the potential user group accordingly. Although internet banking is widely available in Jamaica there have been few studies that have sought to understand the characteristics of its users. This study analyses survey data of internet banking users and non-users using data mining techniques to identify the demographic and behavioural characteristics of internet banking users.

 

 

  1. Business Intelligence for Sustainable Competitive Advantage: Field Study of Telecommunications Industry

Azizah Ahmad, Department of Applied Science, Universiti Utara Malaysia,06010 UUM Sintok & Norshuhada Shiratuddin, Department of Applied Science, Universiti Utara Malaysia, 06010 UUM Sintok.

 

Business Intelligence (BI) as a concept is not a new phenomenon. It has been widely emphasized in the strategic management literature as an essential competitive tool, while IT related literature stresses on technical advancement of BI tools. However, deployment of BI in

organizations, which involve IT systems and other organizational resources to manage knowledge strategically that would lead organizations to sustain their competitive advantage, is not well explained. While the literature on BI covers various issues especially on technical aspects, it lacks comprehensive studies of successful BI deployment and its relationship with organizations sustainable competitive advantage. This paper attempts to highlight these issues

in the context of Telecommunication Industry. A qualitative field study in Malaysia is undertaken in this research, where all of four telecommunication services providers, in various levels of BIdeployments, are studied. The study is conducted via interviews with key personnel, whom are involved in decision-making tasks in their organizations. Contents analysis is then performed to extract the factors and variables and a comprehensive model of BI for Sustainable Competitive Advantage is developed. The results of the interviews identify nine major variables affecting successful BI deployment as; Quality Information, Quality Users, Quality Systems, BI Governance, Business Strategy, Use of BI Tools, and Organization Culture. BI is believed to the main source for acquiring knowledge in sustaining competitive advantage.

 

 

  1. A Link-Based Method for Hiding Sensitive Patterns on Data Mining

Yu-Chiang Li, Computer Science and Information Engineering, Southern Taiwan University,

Yung-Kang, Tainan, 71005, Taiwan & Chien-Ting Huang, Computer Science and Information Engineering, Southern Taiwan University, Yung-Kang, Tainan, 71005, Taiwan

 

Shared data has become a trend among businesses to increase mutual benefit for all participants. However, sharing data also increases the risk of unexpected information leaks, so it is important that an organization develops an efficient and effective method restricting information leaks to maintain its competitive edge. The direct release of databases for query is particularly vulnerable to sensitive information leaks. The challenge is to modify the data in order to protect the sensitive individual information while maintaining the data mining capability. This article proposes a link-based model for effective association rule hiding that reduces the unexpected loss of association rules, which is called the side effect or misses cost. Experimental results reveal that the proposed link-based model is effective and can generally achieve a significantly lower misses cost than the Algo2b and MaxFIA methods do in test datasets.

 

 

  1. System Dynamics Simulation for Customer Lifetime Value

Jiann-Horng Lin, Department of Information Management, I-Shou University, Taiwan & Hui-Ling Hung, Department of Information Management, I-Shou University, Taiwan

 

Relationship marketing assumes that firms can be more profitable if they identify the most profitable customers and invest disproportionate marketing resources in them. Customer Lifetime Value is one of the key metrics in marketing and is considered an important segmentation base. In this paper, we use the system dynamic model as a policy laboratory to test a lot of policies and evolve the appropriate strategy. According to the simulation of the model, we have to improve the three key parameters, included Repeat Purchase Rate, Communication Rate and Marketing Costs in order to increase the revenue.

 

  1. A Brief Review of Recent Research Trends on Applications of Computational and Statistical Techniques in Financial & Business Intelligence

Kuldeep Kumar, School of Business, Bond University, Gold Coast, Australia & Sukanto Bhattacharya, Deakin Business School, Deakin University, Melbourne, Australia.

 

Artificial neural networks and statistical techniques like decision trees, discriminant analysis, logistic regression and survival analysis play a crucial role in Business Intelligence. These predictive analytical tools exploit patterns found in historical data to make predictions about future events. In this paper we have shown some recent developments of a few of these techniques in financial and business intelligence applications like fraud detection, bankruptcy prediction and credit rating scoring.

 

 

  1. Information Management and Accessibility in Business Intelligence in Cloud Vs Conventional Business Intelligence

Somina Venkata Surya Brahma, Linga Sarma, DW & BI Consultant and Architect, Plot No: 54B Road No: 4 H No: 2-3-381 & Sai Nagar, Nagole, Hyderabad-500068, India

 

How effective it would be having the information stored and managed using Business Intelligence in a Cloud Vs Conventional Business Intelligence. Apart from the cost benefits and access to robust runtime environment from a Cloud BI, there are various other issues with respect implementation of Cloud BI. This white paper illustrates and contrasts the Conventional Business Intelligence Vs a Cloud BI. The challenges that are not addressed part of the Cloud BI like even the midst of fear over the loss of Data Control, Security and Privacy and also Cloud poses a problem of BI Architects as scalability would entitle an extra expenditure in hardware and software more complexity for data fungibility and loss in data quality and consistency amongst other critical issues as said above. The other challenges would be in adopting the new skills for implementing BI in a Cloud for the existing BI technology vendors. Also how a Cloud BI would address the existing organizations EAI and EII needs in a supply chain environment. It is more like an effectiveness of implementing any new change and methodology quickly, efficiently and moreover accommodating data volumes on a shared platform. What are the benefits one would get as being operated on a independent platform Vs shared platform and the question of how the above said issues are better handled in a conventional BI process. Case studies are presented to compare and contrast the effective information storage, management, in terms of accessibility and other issues and also on the focus on presentation of the data. This would provide a business case for the existing technology vendors to protest and evaluate to an extent, how all the existing conventional BI platforms cannot be migrated to a Cloud. What are the gaps that are not addressed and leftover of not supporting all the BI solution needs, while migrating to Cloud BI.

 

 

  1. Predicting student’s behavior in education using J48 algorithm analysis tools in WEKA environment

Daya Gupta, Dept of Computer Engineering, Delhi Technological University, Delhi, India & Rajni Jindal, Dept of Computer Engineering, Delhi Technological University, Delhi, India & Vaibhav Verma, Dept of Computer Engineering

Delhi Technological University, Delhi, India & Dilpreet Singh Kohli, Dept of Computer Engineering, Delhi Technological University, Delhi, India & Shashi Kant Sharma, Dept of Computer Engineering, Delhi Technological University, Delhi, India

 

In the given case study we are proposing a scheme of student evaluation that can help the colleges (or other related organizations) to judge the candidate if he/she matches with the offered program .The student’s satisfaction level for that program can be estimated on the basis of his/her general attributes and program’s attributes. The proposal is to construct a decision tree on the bases of historical data through which the general attributes of the student and the particular program can be mapped to the satisfaction level. Construction and refinement of decision tree makes use of J48 Algorithm. Predictions regarding student dropout, transfer or persist can be made using the estimated satisfaction level.

 

  1. Outsourcing data mining tasks to could while preserving customer provacy

Yi-ming Chen, Dept of information management, National Central University, Jhong-li, Taiwan & Yan-hao Chu, Dept of information management, National Central University, Jhong-li, Taiwan

 

Recently, as more and more entreprises are liely to outsourcing time consuming data mining task to the cloud for saving both time and money, it becomas important to ensure rhe customer's privacy because the decentralized nature of cloud services increases the risk of data leakage. Traditional approaches either present the edge mining server from contacting the original customer data or encrypting the queries to protect the custmer's knowledge behind the query logs. Most of these approaches however, steel need to put hugh volume of original customer data in a central data center which may be attacked by malicious users. In this paper, we propose propose a BKM approach to solve this problem. By the BKM approach, before put to the cloud, all of the customer's data are transformed to a string of 1s and 0s by bloom filters. Then these seems non-meaningful strings are grouped by K-means method. Finally, we use the MFTS (Maximal Frequent Term Set) algorithm to find the most representative patterns to customers. To illustrate the application of BKM approach, we use a set of secutirty alerts adopted from the DARPA dataset to show how the BKM mining can help customers to find out the statistics of the attack patterns while revealing none of customer's privacy in the could.

 

 

  1. Impact of Missing Data and Data Imputation on Data Mining

Marvin L. Brown, Department of Computer Information Systems, College of Business, Grambling State University, Grambling, LA 71245, & Chien-Hua Mike Lin, Department of Computer and Information Science, School of Business, Cleveland State University, Cleveland, OH 44114

 

The purpose of this paper is to investigate the impact of missing data and data imputation on the data mining phase of knowledge discovery when neural networks employing an s-Sigmoid Transfer function are utilized. While studies have been conducted independently in the areas knowledge discovery, missing data and data imputation, only a few have integrated all three dimensions. This research explores the impact of data missingness at various increasing levels in KDD (knowledge discovery databases) models that contain various volumes of case frequencies that employ neural networks as the data mining algorithm. Four of the most commonly utilized data imputation methods – Case Deletion, Mean Substitution, Regression Imputation, and Multiple Imputation, are used to determine their effectiveness in dealing with the issue of data missingness.

 

  1. Visualization of Data in Business Intelligence Systems: An Experimental Examination of the Impact of Chartjunk

Peter OʼDonnell, Decision Support Systems Laboratory, Monash University, Melbourne, Australia & Majid Pourdadash, Decision Support Systems Laboratory, Monash University, Melbourne, Australia.

 

The reports and screens of a typical business intelligence application use a variety of impressive looking gauges, charts and graphics to display data, often stylized with 3-dimensional effects. The literature on information visualization suggests that the use of this type of graphic visualization leads to poor decision-making performance. One simple concept that can be used to improve the design of graphic data is the data-ink ratio. Visualizations with a low data-ink ratio contain unnecessary decoration, sometimes called chartjunk, which must be processed before the actual data can be understood. Chartjunk is minimized in visualizations that have a high data-ink ratio [21]. In order to achieve a high data-ink ratio a graphic will be drawn without redundant or unnecessary visual elements making the data displayed easier to understand. This paper reports the results of an experiment that compares the use of high and low data-ink ratio charts for a series of data analysis tasks. The results support the proposition that data analysis is improved if high data-ink ratio charts are used. The paper concludes by arguing that the use of highly decorated and stylized forms of data display in business intelligence systems is harming rather than supporting the decision making of end users.

 

  1. Hierarchy integration in the design of data warehouses

Yasser Hachaichi, Mir@cl Loboratory, ISAAS, B.P.1013, 3018 Sfax, Tunisia & Jamel Feki

Mir@cl Loboratory, FSEGS, B.P.1088, 3018 Sfax, Tunisia

 

Face to their merger and/or collaboration with partners, today’s enterprises often need to integrate several databases. As a result, their decision making process ends up analyzing data coming from various databases. While database integration has been thoroughly examined, it is only recently that the integration of multidimensional models has drawn attention. A multidimensional model is a data model that facilitates the analysis operations during the decision making process. It organizes data into facts that can be analyzed according to dimensions represented through hierarchies. This paper presents an integration process for the hierarchy concept. In particular, it proposes a set of basic integration operations and constraints that produce a multidimensional model that is loadable from different data sources. It illustrates the integration process through examples.

 

  1. Business Intelligence for Financial Information: An Overview of XBRL

Jarupa Vipoopinyo, Department of Electronic and Computer Engineering, University of Portsmouth, United Kingdom & Said Rabah Azzam, Department of Electronic and Computer Engineering, University of Portsmouth, United Kingdom & Shikun Zhou, Department of Electronic and Computer Engineering, University of Portsmouth, United Kingdom.

 

This paper presents the first stage of our proposed research work. In general, our first stage is to identify and review one of the key business reporting languages, namely, Extensible Business Reporting Language (XBRL). XBRL is an emerging standard based on XML technology, particularly to streamline a diversification of financial information throughout the world for globalised stakeholders. This paper gives an overview and discusses potential of using the current use of relevant intelligent technologies for supporting and improving XBRL for a more automated manner and to enhance usability of XBRL. Furthermore, a model of verification and validation program is created to further develop into common proof framework to improve quality of financial information and achieve a success and effectiveness of XBRL. This will lead business users and decision makers to interact with the accurate and reliable information.

A Pattern-Supported Approach for Conceptual Modeling of Dimensional Data Models

Company Description : Dimensional data modeling is an important activity in the development of business intelligence systems, especially when a data warehouse or data marts are to be created. Dimensional data modeling involves identification of required fact and dimension entities that support various types of BI functionality. Since the process of developing dimensional data models differs significantly from that of developing data models for online transaction processing systems, practitioners and researchers have been suggesting approaches to aid this process. Commonly employed approaches for dimensional data modeling make use of operational data models or enterprise data models in identifying grains, dimensions and facts. This paper presents an approach for conceptual modeling of dimensional model with the analysis patterns especially for the analysis of source data models in identifying facts and dimensions, and for the depiction of designed dimensional models. The proposed approach is illustrated for analyzing source data models of an airline company and for developing dimensional data models. The application of analysis patterns is expected to result in an efficient and effective modeling process by facilitating communication among the stakeholders.

Product Type : Academic Conferences

Author : Narasimha Bolloju, City University of Hong Kong, Tat Chee Avenue

PDF 7p

Languages : English

Estimating Consumer Demand From High-Frequency Data

Company Description : Price discrimination offers sellers the possibility of increasing revenue by capturing a market’s consumer surplus: arising from the low price elasticity segment of customers that would have been prepared to pay more than the current market price. First degree price discrimination requires a seller to know the maximum (reserve) price that each consumer is willing to pay. Discouragingly, this information is often unavailable; making the theoretical ideal a practical impossibility. Electronic commerce offers a solution; with loyalty cards, transaction statements and online accounts all providing channels of customer monitoring. The vast amount of data generated–eBay alone produces terabytes daily–creates an invaluable repository of information that, if used intelligently, enables consumer behaviour to be modelled and predicted. Thus, once behavioural models are calibrated, price discrimination can be tailored to the level of the individual. Here, we introduce a statistical method designed to model the behaviour of bidders on eBay to estimate demand functions for individual item classes. Using eBay’s temporal bidding data–arrival times, price, user id–the model generates estimates of individual reserve prices for market participants; including hidden, or censored, demand not directly contained within the underlying data. Market demand is then estimated, enabling eBay power sellers–large professional bulk-sellers–to optimize sales and increase revenue. Proprietary software automates this process: analyzing data; modelling behaviour; estimating demand; and generating sales strategy. This work is a tentative first step of a wider, ongoing, research program to discover a practical methodology for automatically calibrating models of consumers from large-scale high-frequency data. Multi-agent systems and artificial intelligence offer principled approaches to the modelling of complex interactions between multiple individuals. The goal is to dynamically model market interactions using realistic models of individual consumers. Such models offer greater flexibility and insight than static, historical, data analysis alone.

Product Type : Academic Conferences

Author : John Cartlidge∗, University of Central Lancashire, Preston, PR1-

PDF 7p

Languages : English

Conceptualizing a True Business Intelligence System

Company Description : This paper concentrates on the development of a framework to build a true business intelligence system to represent natural intelligence of decision makers in any organization. Further, to incorporate intelligence in a system, a detailed discussion on various factors associated with intelligence are discussed. Also, this paper presents an analysis of the interrelationship among data, information, knowledge, and intelligence. In addition, a classification of knowledge is presented to understand the intelligence building process. Finally, a framework for building a true business intelligence system is provided in line with the development of the adaptive business intelligence system.

Product Type : Academic Conferences

Author : Anjan Kumar Swain, Indian Institute of Management Kozhikode, IIM

PDF 5p

Languages : English

Intranet: How is Business Intelligence Stored in a Data Warehouse?

Company Description : Business intelligence has long been regarded as a critical success factor for contemporary business. While its importance is known, research on what data or information is related to business intelligence and how business intelligence is captured for use in an organizational information system such as a private intranet for use within a wider organizational community is limited. This research discusses how business intelligence is gathered, created, used and realized through the storing in an organizational data warehouse in an intranet. A case study was conducted to examine how an Intranet system was designed in a selected organization, how business intelligence is collected and organized, and how it is used for strategic planning, decision making and daily business operation purposes. This research adopts a qualitative case study method using interviews and observation techniques. The respondents explained how structured business intelligence data was categorized and disseminated to users, and how the used information empowered staff in work performance. It was discovered that the design of the intranet directly relates to retaining knowledge of the staff within the organization, drawing all internal resources together, capturing resources from external sources, and forming a repository of organizational assets for common use embedded through organizational work procedures within the intranet.

Product Type : Academic Conferences

Author : Lisa Soon, School of Information & Communication Technology, Cen

PDF 7p

Languages : English

Log mining for query recommendation in e-commerce

Company Description : Searching is very important in e-commerce application, especially if its content is generated by users. A query recommendation system, which recommends new queries to a user after his initial query, plays a key role for shortening the search session by guiding the user to reach the product they need. Because the amount of potential queries is nearly infinite, query recommendation can be consifered as an information filtering processn which may either be content-based filtering or collaborative filtering. In this paper, we desrcribe hybrid method for performing scalable, fast responding query recommendation system, which is combination content-based filtering and collaborative filtering techniques. We show how the method has been applied to a set of user log data collected from an e-commerce site by creating model in two-monthd of query logs and evaluated the model using separate set of querylogs of one-month period. Finally, since (i) most og yhe query seassions are shhort, (ii) the goal of the query recommendation system is keeping the session short, and (iii) traditional evaluation metrics suffer from the lack of rated data; we introduce a new evaluation metric for query recommendation. The new metric returns closer results to humain-based evaluation measures. Results obtained from the evaluation show that the system decrease the search path, and allows users to reach product with fewer cliks.

Product Type : Academic Conferences

Author : Gokhan Capan, AnadouUniversity & Alper Kursat Uysal, AnadouUniv

PDF 5p

Languages : English

Debugging Support for Data Warehouse Applications

Company Description : The paper describes a model-based approach to developing a general tool for localizing faults in applications of data warehouse technology. A model of the application is configured from a library of generic models of standard (types of) modules and exploited by a consistency-based diagnosis algorithm, originally used for diagnosing physical devices. Observing intermediate results can require high efforts or even be impossible, which limits the discriminability between different faults in a sequence of data processing steps. To compensate for this, fault models are used. This becomes a feasible solution for standard modules of a data warehouse application along with a stratification of the data. Fault models capture the potential impact of faults of process steps and data transfer on the data strata as well as on sets of data. Reflecting the nature of the initial symptoms and of the potential checks, these descriptions are stated at a qualitative level. The solution has been validated in customer report generation of a provider of mobile phone services.

Product Type : Academic Conferences

Author : Peter Struss, Tech.Univ. of Munich, Boltzmannstr. 3, D-85748 Gar

PDF 7p

Languages : English

Identifying the Needs and Benefits of an Electronic Medical Record (EMR) System through Resea

Company Description : India is a developing country with an advanced technical base and skill set. Although the technical growth is large, medical provision to the entire country is still a challenge. The low income population in rural and urban areas is often deprived of organized medical facilities, resulting in improper and inadequate patient care and medical development. To resolve this matter we propose the implementation of an EMR system. An EMR system is “an electronic record of health-related information on an individual that can be created, gathered, managed, and consulted by authorized clinicians and staff within one healthcare organization.”(NAHIT, 2008, 6). This system is an efficient method to organize the healthcare system in the rural areas to provide patient care under monitored, documented and advanced conditions.

Product Type : Academic Conferences

Author : Arshia Khan, Department of Computer Information Systems, The Col

PDF 15p

Languages : English

Using data mining process standards: A case study - Profiling Internet Banking Users in Jamai

Company Description : Internet banking can provide a number of benefits both to the bank and its customers, however, there is a cost associated with providing this service and internet banking will only be financially viable if these costs can be justified. Justifying these costs requires an understanding of the likely customer base that will take up this banking option so that banks can target the potential user group accordingly. Although internet banking is widely available in Jamaica there have been few studies that have sought to understand the characteristics of its users. This study analyses survey data of internet banking users and non-users using data mining techniques to identify the demographic and behavioural characteristics of internet banking users.

Product Type : Academic Conferences

Author : Gunjan Mansingh, The University of the West Indies, Department o

PDF 7p

Languages : English

Business Intelligence for Sustainable Competitive Advantage: Field Study of Telecommunication

Company Description : Business Intelligence (BI) as a concept is not a new phenomenon. It has been widely emphasized in the strategic management literature as an essential competitive tool, while IT related literature stresses on technical advancement of BI tools. However, deployment of BI in organizations, which involve IT systems and other organizational resources to manage knowledge strategically that would lead organizations to sustain their competitive advantage, is not well explained. While the literature on BI covers various issues especially on technical aspects, it lacks comprehensive studies of successful BI deployment and its relationship with organizations sustainable competitive advantage. This paper attempts to highlight these issues in the context of Telecommunication Industry. A qualitative field study in Malaysia is undertaken in this research, where all of four telecommunication services providers, in various levels of BIdeployments, are studied. The study is conducted via interviews with key personnel, whom are involved in decision-making tasks in their organizations. Contents analysis is then performed to extract the factors and variables and a comprehensive model of BI for Sustainable Competitive Advantage is developed. The results of the interviews identify nine major variables affecting successful BI deployment as; Quality Information, Quality Users, Quality Systems, BI Governance, Business Strategy, Use of BI Tools, and Organization Culture. BI is believed to the main source for acquiring knowledge in sustaining competitive advantage.

Product Type : Academic Conferences

Author : Azizah Ahmad, Department of Applied Science, Universiti Utara Ma

PDF 7p

Languages : English

A Link-Based Method for Hiding Sensitive Patterns on Data Mining

Company Description : Shared data has become a trend among businesses to increase mutual benefit for all participants. However, sharing data also increases the risk of unexpected information leaks, so it is important that an organization develops an efficient and effective method restricting information leaks to maintain its competitive edge. The direct release of databases for query is particularly vulnerable to sensitive information leaks. The challenge is to modify the data in order to protect the sensitive individual information while maintaining the data mining capability. This article proposes a link-based model for effective association rule hiding that reduces the unexpected loss of association rules, which is called the side effect or misses cost. Experimental results reveal that the proposed link-based model is effective and can generally achieve a significantly lower misses cost than the Algo2b and MaxFIA methods do in test datasets.

Product Type : Academic Conferences

Author : Yu-Chiang Li, Computer Science and Information Engineering, Sout

PDF 6p

Languages : English

System Dynamics Simulation for Customer Lifetime Value

Company Description : Relationship marketing assumes that firms can be more profitable if they identify the most profitable customers and invest disproportionate marketing resources in them. Customer Lifetime Value is one of the key metrics in marketing and is considered an important segmentation base. In this paper, we use the system dynamic model as a policy laboratory to test a lot of policies and evolve the appropriate strategy. According to the simulation of the model, we have to improve the three key parameters, included Repeat Purchase Rate, Communication Rate and Marketing Costs in order to increase the revenue.

Product Type : Academic Conferences

Author : Jiann-Horng Lin, Department of Information Management, I-Shou Un

PDF 6p

Languages : English

A Brief Review of Recent Research Trends on Applications of Computational and Statistical Tec...

Company Description : Artificial neural networks and statistical techniques like decision trees, discriminant analysis, logistic regression and survival analysis play a crucial role in Business Intelligence. These predictive analytical tools exploit patterns found in historical data to make predictions about future events. In this paper we have shown some recent developments of a few of these techniques in financial and business intelligence applications like fraud detection, bankruptcy prediction and credit rating scoring.

Product Type : Academic Conferences

Author : Kuldeep Kumar, School of Business, Bond University, Gold Coast,

PDF 8p

Languages : English

Information Management and Accessibility in Business Intelligence in Cloud Vs Conventional BI

Company Description : How effective it would be having the information stored and managed using Business Intelligence in a Cloud Vs Conventional Business Intelligence. Apart from the cost benefits and access to robust runtime environment from a Cloud BI, there are various other issues with respect implementation of Cloud BI. This white paper illustrates and contrasts the Conventional Business Intelligence Vs a Cloud BI. The challenges that are not addressed part of the Cloud BI like even the midst of fear over the loss of Data Control, Security and Privacy and also Cloud poses a problem of BI Architects as scalability would entitle an extra expenditure in hardware and software more complexity for data fungibility and loss in data quality and consistency amongst other critical issues as said above. The other challenges would be in adopting the new skills for implementing BI in a Cloud for the existing BI technology vendors. Also how a Cloud BI would address the existing organizations EAI and EII needs in a supply chain environment. It is more like an effectiveness of implementing any new change and methodology quickly, efficiently and moreover accommodating data volumes on a shared platform. What are the benefits one would get as being operated on a independent platform Vs shared platform and the question of how the above said issues are better handled in a conventional BI process. Case studies are presented to compare and contrast the effective information storage, management, in terms of accessibility and other issues and also on the focus on presentation of the data. This would provide a business case for the existing technology vendors to protest and evaluate to an extent, how all the existing conventional BI platforms cannot be migrated to a Cloud. What are the gaps that are not addressed and leftover of not supporting all the BI solution needs, while migrating to Cloud BI.

Product Type : Academic Conferences

Author : Somina Venkata Surya Brahma, Linga Sarma, DW & BI Consultant and

PDF 12p

Languages : English

Predicting student’s behavior in education using J48 algorithm analysis tools in WEKA environ...

Company Description : In the given case study we are proposing a scheme of student evaluation that can help the colleges (or other related organizations) to judge the candidate if he/she matches with the offered program .The student’s satisfaction level for that program can be estimated on the basis of his/her general attributes and program’s attributes. The proposal is to construct a decision tree on the bases of historical data through which the general attributes of the student and the particular program can be mapped to the satisfaction level. Construction and refinement of decision tree makes use of J48 Algorithm. Predictions regarding student dropout, transfer or persist can be made using the estimated satisfaction level.

Product Type : Academic Conferences

Author : Daya Gupta, Dept of Computer Engineering, Delhi Technological Un

PDF 6p

Languages : English

Outsourcing data mining tasks to could while preserving customer provacy

Company Description : Recently, as more and more entreprises are liely to outsourcing time consuming data mining task to the cloud for saving both time and money, it becomas important to ensure rhe customer's privacy because the decentralized nature of cloud services increases the risk of data leakage. Traditional approaches either present the edge mining server from contacting the original customer data or encrypting the queries to protect the custmer's knowledge behind the query logs. Most of these approaches however, steel need to put hugh volume of original customer data in a central data center which may be attacked by malicious users. In this paper, we propose propose a BKM approach to solve this problem. By the BKM approach, before put to the cloud, all of the customer's data are transformed to a string of 1s and 0s by bloom filters. Then these seems non-meaningful strings are grouped by K-means method. Finally, we use the MFTS (Maximal Frequent Term Set) algorithm to find the most representative patterns to customers. To illustrate the application of BKM approach, we use a set of secutirty alerts adopted from the DARPA dataset to show how the BKM mining can help customers to find out the statistics of the attack patterns while revealing none of customer's privacy in the could.

Product Type : Academic Conferences

Author : Yi-ming Chen, Dept of information management, National Central U

PDF 7p

Languages : English

Impact of Missing Data and Data Imputation on Data Mining

Company Description : The purpose of this paper is to investigate the impact of missing data and data imputation on the data mining phase of knowledge discovery when neural networks employing an s-Sigmoid Transfer function are utilized. While studies have been conducted independently in the areas knowledge discovery, missing data and data imputation, only a few have integrated all three dimensions. This research explores the impact of data missingness at various increasing levels in KDD (knowledge discovery databases) models that contain various volumes of case frequencies that employ neural networks as the data mining algorithm. Four of the most commonly utilized data imputation methods – Case Deletion, Mean Substitution, Regression Imputation, and Multiple Imputation, are used to determine their effectiveness in dealing with the issue of data missingness.

Product Type : Academic Conferences

Author : Marvin L. Brown, Department of Computer Information Systems, Col

PDF 10p

Languages : English

Visualization of Data in Business Intelligence Systems: An Experimental Examination of the Im...

Company Description : The reports and screens of a typical business intelligence application use a variety of impressive looking gauges, charts and graphics to display data, often stylized with 3-dimensional effects. The literature on information visualization suggests that the use of this type of graphic visualization leads to poor decision-making performance. One simple concept that can be used to improve the design of graphic data is the data-ink ratio. Visualizations with a low data-ink ratio contain unnecessary decoration, sometimes called chartjunk, which must be processed before the actual data can be understood. Chartjunk is minimized in visualizations that have a high data-ink ratio [21]. In order to achieve a high data-ink ratio a graphic will be drawn without redundant or unnecessary visual elements making the data displayed easier to understand. This paper reports the results of an experiment that compares the use of high and low data-ink ratio charts for a series of data analysis tasks. The results support the proposition that data analysis is improved if high data-ink ratio charts are used. The paper concludes by arguing that the use of highly decorated and stylized forms of data display in business intelligence systems is harming rather than supporting the decision making of end users.

Product Type : Academic Conferences

Author : Peter OʼDonnell, Decision Support Systems Laboratory, Monash Uni

PDF 8p

Languages : English

Hierarchy integration in the design of data warehouses

Company Description : Face to their merger and/or collaboration with partners, today’s enterprises often need to integrate several databases. As a result, their decision making process ends up analyzing data coming from various databases. While database integration has been thoroughly examined, it is only recently that the integration of multidimensional models has drawn attention. A multidimensional model is a data model that facilitates the analysis operations during the decision making process. It organizes data into facts that can be analyzed according to dimensions represented through hierarchies. This paper presents an integration process for the hierarchy concept. In particular, it proposes a set of basic integration operations and constraints that produce a multidimensional model that is loadable from different data sources. It illustrates the integration process through examples.

Product Type : Academic Conferences

Author : Yasser Hachaichi, Mir@cl Loboratory, ISAAS, B.P.1013, 3018 Sfax,

PDF 9p

Languages : English

Business Intelligence for Financial Information: An Overview of XBRL

Company Description : This paper presents the first stage of our proposed research work. In general, our first stage is to identify and review one of the key business reporting languages, namely, Extensible Business Reporting Language (XBRL). XBRL is an emerging standard based on XML technology, particularly to streamline a diversification of financial information throughout the world for globalised stakeholders. This paper gives an overview and discusses potential of using the current use of relevant intelligent technologies for supporting and improving XBRL for a more automated manner and to enhance usability of XBRL. Furthermore, a model of verification and validation program is created to further develop into common proof framework to improve quality of financial information and achieve a success and effectiveness of XBRL. This will lead business users and decision makers to interact with the accurate and reliable information.

Product Type : Academic Conferences

Author : Jarupa Vipoopinyo, Department of Electronic and Computer Enginee

PDF 5p

Languages : English

Organizer : Global Science & Technology Forum

GSTF provides a global intellectual platform for top notch academics and industry professionals to actively interact and share their groundbreaking research achievements. GSTF is dedicated to promoting research and development and offers an inter-disciplinary intellectual platform for leading scientists, researchers, academics and industry professionals across Asia Pacific to actively consult, network and collaborate with their counterparts across the globe.