Evaluating quality of text clustering with ART1

Massey, Louis

doi:10.1109/ijcnn.2003.1223901

Cited by 9 publications

(19 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 2 shows the set of training parameters for the conventional ART1 using the method in (Massey, 2003) and the three variations of ART1-P. Clearly, no rule can be deduced from the empirical results; the ART1-P selected a higher ρ and smaller a factors, while the maximum number of clusters came more realistic with the ART1-P in the sense of expert knowledge. Table 3 shows the performance of clustering using a test set of unseen 1500 input patterns, with three variations of the proposed method against the conventional one based on a recommended set of training parameters.…”

Section: Resultsmentioning

confidence: 99%

“…We applied a feature extraction procedure similar to (Lippmann, 1987;Massey, 2002;Massey, 2003;Cohen et al, 2005;Davidov et al, 2004;University of Waikato, 2005), to produce training and testing data sets. Each web page is represented by a string with a set of 2000 useful tokens with threshold frequencies of 3.…”

Section: Methodsmentioning

confidence: 99%

“…We compare the ART1-Px performance with that of the conventional ART1 with a recommended set based on the work of (Lippmann, 1987;Massey, 2002;Massey, 2003). The work will not be compared to that of ART1-GEP in (Al-Natsheh and Eldos, 2007) for few reasons; the work team has split and we have no access to the code, the data set has been augmented based on recommendations from some reviewers and the target of this work is to select a pretty good set of parameters at a much lower cost in comparison with the ART1-GEP.…”

Section: Methodsmentioning

confidence: 99%

“…The range for each of these parameters is shown in Table 1, where the minimal vigilance computed as ρ min = 1/N, where N is the number of features (tokens) used to represent a document (Massey, 2003), which is 2000 in our tests. Because the repetitive interval halving considers the middle of the halves, we would start with 0 for ρ, so when it gets portioned many times it would be allowed to assume values as low as 1/N.…”

Section: Methodsmentioning

confidence: 99%

“…Our guide to select a good set of parameters must be the performance metrics used at the end to judge the quality of clustering, which are the well known (FM) and (JAC) measures (Massey, 2003), defined as: However, those measures do not reflect the parameter selection robustness; the impact of training set ordering on the performance. Practically, we are interested in a set of parameters that would give better performance regardless of the data set training order.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Adaptive Resonance Theory Training Parameters: Pretty Good Sets

Eldos¹

2010

Journal of Computer Science

View full text Add to dashboard Cite

Problem statement: ART1 artificial neural networks offer good tools for test clustering, where no expert is needed if the system is well trained. However, having no output reference for the input patterns makes it hard to judge the quality of the training. Moreover, the performance depends to a great extent on a set of training parameters. Designers follow some recommendations or depend on their expertise in finding good sets with no performance guarantees. Many methods were proposed; from greedy methods offering quick and acceptable solutions to evolutionary algorithms offering suboptimal sets of parameters. While the evolutionary algorithms are a good choice for quality, the computational cost is large even for an offline process; after all, computing resources are not for free. Approach: We introduced a method for selecting a set of parameters that yields a comparable performance and robust operation, with relatively low cost compared to the evolutionary methods. This method located a suitable set through repetitive portioning of the range, by considering the best subset for the next iteration. Results: Tests have shown that performance comparable with the computationally intensive evolutionary methods could be achieved in much less time. Conclusion: The repetitive portioning method for finding a good set of training parameters is very cost effective and yields good performance.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Adaptive Resonance Theory Training Parameters: Pretty Good Sets

Eldos¹

2010

Journal of Computer Science

View full text Add to dashboard Cite

show abstract

An Improved ART1 neural network algorithm for character recognition

2010

2010 Chinese Control and Decision Conference

View full text Add to dashboard Cite

Adaptive Resonance Theory, or ART, was introduced as a theory of human cognitive information processing [1]- [2]. Its inspiration is neurobiological and its component parts are intended to model a variety of hierarchical inference levels in the human brain. ARTneural network that includes ART1, ART2 and ART3 is an unsupervised learning method. The model develops a theory for neural networks for learning pattern categories stably in response to inputs. The model is intended to solve the stability-plasticity dilemma that is faced by all the dominant intelligent systems using neural network paradigms. The properties of plasticity and stability are intimately related. A characteristic of an adequate system is the ability to adaptively switch between its stable and plastic modes. It shall be capable of maintaining plasticity in order to learn about significant new events, while remaining stable in response to irrelevant or often repeated events. ARTl network model is a self-organizing architecture capable of learning recognition categories of complex binary input patterns. It is expected to be capable of distinguishing between familiar and unfamiliar events, as well as between expected and unexpected events. The advantages of using an ART1 neural network for pattern recognition are tremendous. ART1 algorithm supports on-line learning, allowing new exemplars or events to be immediately clustering. The model doesn't have to be modified to develop a new level of optimization. The architecture of ART1 network is also structured such that numerous highly connected nodes each operate in a similar manner, making these nodes ideal for a parallel implementation. As illustrated, incorporation of the ART1 network into a parallel environment can be done with relative ease, due to the inherent parallel nature of the algorithm.Abstract: The paper indicates the shortage of standard ART1 neural network, and an improved calculating method of similarity is presented. The corresponding place value of two vectors at the same time is considered in this method. The method avoids the different result of ART1 neural network because of inputting different sequence. In order to solve the pattern excursion problem of ART1 neural network, the principle of minority subordinate to majority is proposed to reduce the appeared problem. They improve the applicative effect of ART1 neural network.

show abstract

Commercializing computational intelligence techniques in a business intelligence application

Al-Natsheh¹,

Zalzala

2010

IEEE Congress on Evolutionary Computation

View full text Add to dashboard Cite

Evaluating quality of text clustering with ART1

Cited by 9 publications

References 22 publications

Adaptive Resonance Theory Training Parameters: Pretty Good Sets

Adaptive Resonance Theory Training Parameters: Pretty Good Sets

An Improved ART1 neural network algorithm for character recognition

Commercializing computational intelligence techniques in a business intelligence application

Contact Info

Product

Resources

About