Actionable Knowledge Discovery

Actionable Knowledge Discovery

Longbing Cao (University of Technology Sydney, Australia)
DOI: 10.4018/978-1-60566-026-4.ch002
OnDemand PDF Download:
$37.50

Abstract

Actionable knowledge discovery is selected as one of the greatest challenges (Ankerst, 2002; Fayyad, Shapiro, & Uthurusamy, 2003) of next-generation knowledge discovery in database (KDD) studies (Han & Kamber, 2006). In the existing data mining, often mined patterns are nonactionable to real user needs. To enhance knowledge actionability, domain-related social intelligence is substantially essential (Cao et al., 2006b). The involvement of domain-related social intelligence into data mining leads to domaindriven data mining (Cao & Zhang, 2006a, 2007a), which complements traditional data-centered mining methodology. Domain-related social intelligence consists of intelligence of human, domain, environment, society and cyberspace, which complements data intelligence. The extension of KDD toward domain-driven data mining involves many challenging but promising research and development issues in KDD. Studies in regard to these issues may promote the paradigm shift of KDD from data-centered interesting pattern mining to domain-driven actionable knowledge discovery, and the deployment shift from simulated data set-based to real-life data and business environment-oriented as widely predicted.
Chapter Preview
Top

Background

In the last decades, data mining, or KDD, has become a prominent, exciting research and development area in the field of information technology. Data-centered data mining has experienced rapid development in various aspects such as data mined, knowledge discovered, techniques developed, and applications involved. Table 1 illustrates such key research and development progress in KDD.

Table 1.
An overview of data mining
DimensionKey research progress
Data minedRelational, transactional, object-relational, active, temporal, spatial, time-series, heterogeneous, legacy, Web, and so forth.
Stream, spatiotemporal, multimedia, ontology, event, activity, link, graph, text, sensor, and so forth.
Techniques studiedDatabase, machine learning, or statistics-oriented, say Neural Network, Bayesian network, Support Vector Machine, Rough Set, and so forth.
Association, frequent pattern analysis, multidimensional and OLAP analysis methods, classification, cluster analysis, outlier detection, visualization, and so forth.
Scalable data mining, stream data mining, spatiotemporal data mining, multimedia data mining, biological data mining, text and Web mining, privacy-preserving data mining, event mining, link mining, ontology mining, granule mining, and so forth.
Knowledge discoveredCharacters, associations, classes, clusters, discrimination, trends, deviation, outliers, exceptions and so forth.
Application involvedEngineering, retail market, telecommunication, banking, fraud detection, intrusion detection, stock market, social security, bio-informatics, defense, Web services, biological, social network analysis, intelligence and security, and so forth.
Enterprise data mining, cross-organization mining, online mining, dynamic mining, and so forth.

Key Terms in this Chapter

Business Expectation: A pattern of interest to business needs satisfies business performance requests from aspects such as social, economic and psychoanalytic concerns.

Domain Driven Data Mining: Data mining methodologies and techniques that utilize domain-oriented social intelligence, target dependable, trustworthy and actionable knowledge for business decision making.

Statistical Significance: A pattern mined is statistically significant based on statistical metrics that measure and evaluate the performance of an identified pattern.

Interestingness: Measuring the performance of discovered knowledge in terms of statistical significance, user preference, and business expectation from objective and subjective perspectives.

Intelligence Metasynthesis: The interaction and integration between multiple types of intelligence surrounding a problem solving process, including human intelligence, data intelligence, domain intelligence and cyberspace intelligence. Advanced intelligence emerges through intelligence interaction and metasynthesis.

Social Intelligence: Intelligence hidden or reflected in aspects of data, domain, human, society and cyberspace where appropriate, for instance, domain-specific background information and knowledge, expertise, expert involvement, constraints, environment, business rules and processes.

In-Depth Mining: Mining patterns that disclose deep hidden information and relationship of attributes, which can assist deeper understanding of data, business and decision-making.

Business Decision Making: Decisions made by business people aim to the achievement of business objectives in a manner of satisfying business needs and expectations.

Complete Chapter List

Search this Book:
Reset