Mining Patterns with Domain Knowledge: A Case Study on Multi-language Data

Antunes, Cláudia; Bebiano, Tiago

doi:10.1109/icis.2012.70

Cited by 3 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent work (Antunes and Bebiano 2012) has demonstrated that constraints defined over ontologies may be used by constrained adaptations of the most-well known algorithms for pattern mining, namely apriori (Agrawal and Srikant 1994) and FP-growth (Han, Pei and Yin 2000), without impairing their efficiency, but enabling the reduction of the number of patterns discovered, focusing the discovery according to user expectations.…”

Section: Knowledge Explorationmentioning

confidence: 99%

New Trends in Knowledge Driven Data Mining

Antunes

Silva

2014

Proceedings of the 16th International Conference on Enterprise Information Systems

View full text Add to dashboard Cite

Existing mining algorithms, from classification to pattern mining, reached considerable levels of efficiency, and their extension to deal with more demanding data, such as data streams and big data, show their incontestable quality and adequacy to the problem. Despite their efficiency, their effectiveness on identifying useful information is somehow impaired, not allowing for making use of existing domain knowledge to focus the discovery. The use of this knowledge can bring significant benefits to data mining applications, by resulting in simpler and more interesting and usable models. However, most of existing approaches are concerned with being able to mine specific domains, and therefore are not easily reusable, instead of building general algorithms that are able to incorporate domain knowledge, independently of the domain. In our opinion, this requires a drift in the focus of the research in data mining, and we argue this change should be from domain-driven to knowledge-driven data mining, aiming for a stronger emphasis on the exploration of existing domain knowledge for guiding existing algorithms.

show abstract

Section: Knowledge Explorationmentioning

confidence: 99%

New Trends in Knowledge Driven Data Mining

Antunes

Silva

2014

Proceedings of the 16th International Conference on Enterprise Information Systems

View full text Add to dashboard Cite

show abstract

“…The D 2 PM framework [15] has the goal of supporting the process of pattern mining with the use of domain knowledge, represented through a domain ontology, and encompasses the definition of pattern mining methods able to discover transactional and sequential patterns, guided by constraints. These constraints are the core of this framework, since they are the responsible for incorporating existing knowledge in the mining algorithm.…”

Section: Temporal Constraints and The D 2 Pm Frameworkmentioning

confidence: 99%

(TD)2PaM: A Constraint-Based Algorithm for Mining Temporal Patterns in Transactional Databases

Pina

Antunes

2013

Progress in Artificial Intelligence

View full text Add to dashboard Cite

The analysis of frequent behaviors regarding temporal issues begins to achieve some interest, in particular in the area of health care. However, existing approaches tend to ignore the temporal information and only make use of the order among events occurrence. In this paper, we introduce the notion of temporal constraint, and propose three instantiations of it: complete cyclic temporal constraints, partial cyclic temporal constraints and timespan constraints. Additionally, we propose a new algorithm-(TD) 2 PaM, that together with these constraints, makes possible to focus the pattern mining process on looking for cyclic and timespan patterns. Experimental results reveal the algorithm to be as efficient as its predecessors, and able to discover more informed patterns.

show abstract