An error-based conceptual clustering method for providing approximate query answers

Chu, Wesley W.; Chiang, Kuorong; Hsu, Chih‐Cheng; Yau, Henrick

doi:10.1145/272682.272714

Cited by 21 publications

(15 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…10 However, BinaryCut(C) can handle only ordered data, which is not applicable to M Os. To address this, we replace BinaryCut(C) with k-Means algorithm (k = 2) to find the best cut of M Os without losing its originality.…”

Section: Resultsmentioning

confidence: 99%

“…To incorporate numeric values, some extensions of the COBWEB system are proposed, such as COBWEB/3, 6 ECOBWEB, 7 AUTOCLASS, 8 Generality-based conceptual clustering (GCC), 9 and Error-based conceptual clustering (ECC). 10 However, all of the methods mentioned above do not represent the relations among the clusters or the appropriate meaning of each cluster. Also, they are not very suitable for moving object databases since video data have spatial and temporal characteristics, and high-dimensional attributes.…”

Section: -4mentioning

confidence: 99%

See 1 more Smart Citation

A model-based conceptual clustering of moving objects in video surveillance

Lee¹,

Rajauria²,

Shah³

2007

SPIE Proceedings

View full text Add to dashboard Cite

Data mining techniques have been applied in video databases to identify various patterns or groups. Clustering analysis is used to find the patterns and groups of moving objects in video surveillance systems. Most existing methods for the clustering focus on finding the optimum of overall partitioning. However, these approaches cannot provide meaningful descriptions of the clusters. Also, they are not very suitable for moving object databases since video data have spatial and temporal characteristics, and high-dimensional attributes. In this paper, we propose a model-based conceptual clustering (MCC) of moving objects in video surveillance based on a formal concept analysis. Our proposed MCC consists of three steps: 'model formation', 'model-based concept analysis', and 'concept graph generation'. The generated concept graph provides conceptual descriptions of moving objects. In order to assess the proposed approach, we conduct comprehensive experiments with artificial and real video surveillance data sets. The experimental results indicate that our MCC dominates two other methods, i.e., generality-based and error-based conceptual clustering algorithms, in terms of quality of concepts.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: -4mentioning

confidence: 99%

A model-based conceptual clustering of moving objects in video surveillance

Lee¹,

Rajauria²,

Shah³

2007

SPIE Proceedings

View full text Add to dashboard Cite

show abstract

“…Another important issue in video database management is video data mining to find the knowledge in it. Specifically, finding concepts from video data can solve the problem of unstructured data format, such as conceptual clustering algorithm [22,23,24,25,26,27]. The final goal of video database management is how to retrieve video data efficiently and effectively.…”

Section: Organizationmentioning

confidence: 99%

“…In other words, it cannot be used for abstracting numeric data. In order to incorporate numeric values, some extensions of the COBWEB system are proposed, such as COBWEB/3 [23], ECOBWEB [24], AUTOCLASS [25], Generality-based conceptual clustering (GCC) [26], and Error-based conceptual clustering (ECC) [27]. COBWEB/3…”

Section: Video Data Miningmentioning

confidence: 99%

A Graph-Based Approach for Modeling and Indexing Video Data

Lee

2006

Eighth IEEE International Symposium on Multimedia (ISM'06)

View full text Add to dashboard Cite

“…Clustering algorithms have been developed to generate TAHs automatically from data sources based on a set of attributes selected by the user [CC94,MC93,CCHY96]. Therefore, the TAHs are customized based on the user and context.…”

Section: Query Relaxation Techniquesmentioning

confidence: 99%

Cooperative and Reactive Agents for Information Systems

Chu¹

2000

View full text Add to dashboard Cite

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, 12b. DISTRIBUTION CODE ABSTRACT (Maximum 200 words)Active database systems have received increasing interest from both research and industrial communities. However, trigger conditions in the active rules are often difficult to specify, especially as the complexity of the events and/or conditions increases. To remedy this problem, we propose an active database system that supports rules with conceptual terms, approximate operators, and complex events. The conceptual terms and approximate operators are user and context sensitive. By introducing these high-level constructs, we not only simplify the rule specification process, but also increase the rule expressiveness. Knowledge-based relaxation techniques are used for rule specification and relaxation. High-level concepts and approximate operators used in rules are first relaxed into low-level active rules by using a tree-type knowledge structure called Type Abstraction Hierarchy which can be generated automatically from the database using clustering algorithms. The low-level rules are decomposed into a set of database triggers, which are then submitted into commercial active relational databases for simple trigger-processing. Thus, our proposed high-level active database system supports complex event detection without modification of the underlying database systems. The proposed active database system with high-level rule processing and complex event detection has been implemented at UCLA. The system operates on top of commercial relational databases that demonstrate the feasibility of high-level rule processing and complex event detection. SUBJECT TERMS AbstractActive database systems have received increasing interest from both research and industrial communities. However, trigger conditions in the active rules are often difficult to specify, especially as the complexity of the events and/or conditions increases. To remedy this problem, we propose an active database system that supports rules with conceptual terms, approximate operators, and complex events. The conceptual terms and approximate operators are user and context sensitive. By introducing these highlevel constructs, we not only simplify the rule specification process, but also increase the rule expressiveness. Knowledge-based relaxation techniques are used for rule specification and relaxation. High-level concepts and approximate operators used in rules are first relaxed into low-level active rules by using a tree-type knowledge structure called Type Abstraction Hierarchy which can be generated automatically from the database using clustering algorithms. The low-level rules...

show abstract

An error-based conceptual clustering method for providing approximate query answers

Cited by 21 publications

References 16 publications

A model-based conceptual clustering of moving objects in video surveillance

A model-based conceptual clustering of moving objects in video surveillance

A Graph-Based Approach for Modeling and Indexing Video Data

Cooperative and Reactive Agents for Information Systems

Contact Info

Product

Resources

About