Amit Rudra

Key issues in achieving data quality and consistency in data warehousing among large organisations in Australia

This paper discusses the emergent key issues of data quality in a data warehousing environment. The research study leading to our outcome is described. We investigate the relationship between data quality and data consistency; determine whether data inconsistencies are present in data warehouses and explore how organization ensure, plan and maintain data quality. Our research outcome-an improved understanding of how organizations can maintain data quality and consistency. We found that the quality of data in a data warehouse could be influenced by factors like: data not fully captured, heterogeneous system integration and lack of policy and planning from management.

Data warehouse structuring methodologies for efficient mining of Western Australian petroleum data sources

Representing the knowledge domain of a petroleum system is a complex problem. In the present study, logical modelling of shared attributes of resources industry entities (dimensions or objects) has been used for construction of a dynamic and time-variant metadata model. This work demonstrates effectiveness of multidimensional data modelling for petroleum industry, which will be further investigated for fine-grain data presentation and interpretation for quality knowledge discovery.

Efficiently mining frequent patterns from dense datasets using a cluster of computers

Efficient mining of frequent patterns from large databases has been an active area of research since it is the most expensive step in association rules mining. In this paper, we present an algorithm for finding complete frequent patterns from very large dense datasets in a cluster environment. The data needs to be distributed to the nodes of the cluster only once and the mining can be performed in parallel many times with different parameter settings for minimum support. The algorithm is based on a master-slave scheme where a coordinator controls the data parallel programs running on a number of nodes of the cluster. The parallel program was executed on a cluster of Alpha SMPs. The performance of the algorithm was studied on small and large dense datasets. We report the results of the experiments that show both speed up and scale up of our algorithm along with our conclusions and pointers for further work.

Teaching Business Process Concepts Using Enterprise Systems in a Globalized Context

There is significant growth in global business operations and ERP systems are deemed as important applications in order to unify and improve fragmented and globalised markets. These systems are complex and students struggle to grasp not only the underlying business concepts but also the technology involved. This paper describes the design and implementation of teaching materials for a multinational cross-company collaboration assignment using a commercial ERP system. It includes elements of i) integrated business processes, ii) globalization and iii) experiential learning in Masters courses at two universities - one in Australia and the other in Norway. We discuss the lessons learned from the pilot study including the development of an evaluation tool, inter-university student communication, evaluation of the learning outcome, and the benefits of the cross-country business role play exercise.

Ontology of Western Australian petroleum data for effective data warehouse design and data mining

Volumes of petroleum resources data are archived in different companies. These are of heterogeneous form - either in relational, hierarchical or network structures. With the widespread use of these databases and explosive growth in their sizes, petroleum businesses face a problem of information overload. Effectively utilizing these data volumes is a major challenge for this type of industry. Data search becomes tedious, at times when specific queries are made, due to data accumulated in several geographic locations. In this research work, we revisit these data and propose to simplify these heterogeneous data structures through an ontological data modelling approach particularly to address the issues of data integration and information sharing. Various ontological models have been described in the context of Western Australian petroleum systems. An ontological framework, with a mechanism to integrate petroleum resources data in a warehouse environment has been investigated. Ontology design benefits and its impacts on designing data mining algorithms are discussed. Ontology approach ensures petroleum data validity that supports petroleum knowledge mining process.

Knowledge Base Smarter Articulations for the Open Directory Project in a Sustainable Digital Ecosystem

We examine the volumes and varieties of data sources of the Open Directory Project (ODP), which can endure, regenerate and flourish with new knowledge. The ODP motivates us in building a knowledge base smarter multidimensional data constructs and models. We articulate the models with new artefacts, addressing the heterogeneity and multidimensionality of the data. The conceptualization and contextualization of various entities and dimensions have emerged with innovation that led us to develop a digital ecosystem-based inventory. The ODP based domain ontologies support the warehouse repository, which accommodates multidimensional data relationships. The concept of a digital ecosystem in the ODP context is to bring the dimensions together and unite with multidimensional schemas. We explore the Big Data, incorporating their characteristics in the ODP constructs and models. The volumes and varieties of the ODP data are logically organized and integrated in the warehouse repositories. The multidimensional data modelling makes the ODP more smart and flexible in an environment, where varieties of business rules and constraints change rapidly. The visualization and interpretation are the other artefacts of the Big Data facilitating us use, reuse, test the interoperability and effectiveness of the data models for sustainable ODP digital ecosystem. We compute the polynomial regressions, based on the data fluctuations of the ODP as observed in the scatter plots, providing new data mining models for knowledge interpretation.

Roles of Multidimensionality and Granularity in Warehousing Australian Resources Data

Granularity of data modeled in multidimensional data structures is an important factor for a data warehouse. Grain sizes and number of dimensions participating in the model are critical in ascertaining the quality of analytical queries that are run on such data warehouses. In this paper, exploration and production data of Australian resources industry, pertinent to oil and gas, over the past five decades have been examined for multidimensionality and grain size. This research shows how using an ER approach combined with multidimensional data modeling helps in considerable reduction in the size of the data warehouse, making it more effective and efficient.

Managing the Embedded Digital Ecosystems (EDE) using Big Data Paradigm

Big data sources and their mining from multitude of ecosystems have been the focus of many researchers in both commercial and research organizations. The authors in the current research have focused on embedded ecosystems with big data motivation. Embedded systems hold volumes and a variety of heterogeneous, multidimensional data, and their sources complicate their organization, accessibility, presentation, and interpretation. Objectives of the current research are to provide improved understanding of ecosystems and their inherent connectivity by integrating multiple ecosystems’ big data sources in a data warehouse environment and their analysis with multivariate attribute instances and magnitudes. Domain ontologies are described for connectivity, effective data integration, and mining of embedded ecosystems. The authors attempt to exploit the impacts of disease and environment ecosystems on human ecosystems. To this extent, data patterns, trends, and correlations hidden among big data sources of embedded ecosystems are analyzed for domain knowledge. Data structures and implementation models deduced in the current work can guide the researchers of health care, welfare, and environment for forecasting of resources and managing information systems that involve with big data. Analyzing embedded ecosystems with robust methodologies facilitates the researchers to explore scope and new opportunities in the domain research.

Mining value-based item packages – an integer programming approach

Traditional methods for discovering frequent patterns from large databases assume equal weights for all items of the database. In the real world, managerial decisions are based on economic values attached to the item sets. In this paper, we first introduce the concept of the value based frequent item packages problems. Then we provide an integer linear programming (ILP) model for value based optimization problems in the context of transaction data. The specific problem discussed in this paper is to find an optimal set of item packages (or item sets making up the whole transaction) that returns maximum profit to the organization under some limited resources. The specification of this problem allows us to solve a number of practical decision problems, by applying the existing and new ILP solution techniques. The model has been implemented and tested with real life retail data. The test results are reported in the paper.

Decision-Making Supported by Virtual-World Systems Vis-à-Vis Enterprise Systems’ Uncertainty and Equivocality

Business decision-making is based on a combination of explicit and tacit knowledge. Tacit knowledge is difficult to codify, manage and share in the same manner as explicit knowledge. The ability to share tacit knowledge offers great value to an organisation. The growing level of global business activities leads to an increasingly competitive environment with a greater demand for communication of tacit knowledge over distance. This chapter presents a study of the suitability of virtual-worlds for business decision-making over distance involving the exchange of tacit knowledge to support decisions at the tactical and strategic management layers. The avatar, being the primary interaction object in virtual-worlds, forms the basis of a model of the role concept in user-to-user communication. We present a case study showing how the selection of an avatar reflects the user’s role, and how an avatar situates the user in a business environment. Our study shows that virtual-worlds support the mediation of tacit knowledge. This is contrasted with state-of-the-art enterprise business systems commonly used for business transactions based on explicit knowledge exchanged over distance. The avatar is found to fulfil the role of the communicator and the role expectations from real-life situations. Our findings show that virtual-worlds have unique properties in supporting tacit knowledge exchange required for decision-making in a distributed business environment.