Real-world text clustering with adaptive resonance theory neural networks

Massey, Louis

doi:10.1109/ijcnn.2005.1556360

Cited by 13 publications

(11 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Vigilance values for each level were selected among the best qualities obtained during previous experiments with flat clustering, thus there is no surprise in terms of quality as we choose the highest quality solutions as building blocks for the hierarchy. However, within a real application, a fundamental problem would emerge: quality being unknown at first, the a priori choice of vigilance becomes problematic, as we have pointed out in other work [13]. For k-means and the HAC algorithm implementing the minimum variances criterion, we make two observations.…”

Section: Results With Independent Art1 Networkmentioning

confidence: 99%

“…In fact, the magnitude of the prototypes ||t|| monotonically decreases with time in a ART1 network following updates by intersection: t' = t ∧ x (13) Thus, the new prototype t' is the intersection of the current prototype t and the input x with which it has passed the vigilance test. This means that: || t' || ≤ || t || (14) For example, if t = [1 0 1 1] and x = [1 1 0 0], then ||t|| = 3, ||x|| = 2 and ||t'|| = ||t ∧ x|| = 1, which will necessarily be smaller or equal to ||t||.…”

Section: Results With Art1 Network In Seriesmentioning

confidence: 99%

See 1 more Smart Citation

Discovery of hierarchical thematic structure in text collections with adaptive resonance theory

Massey

2008

Neural Comput Applic

View full text Add to dashboard Cite

This paper investigates the abilities of Adaptive Resonance Theory (ART) neural networks as miners of hierarchical thematic structure in text collections. We present experimental results with binary ART1 on the benchmark Reuter-21578 corpus. Using both quantitative evaluation with the standard F 1 measure and qualitative visualization of the hierarchy obtained with ART, we discuss how useful ART built hierarchies would be to a user intending to use it as a means to find and access textual information. Our F 1 results show that ART1 produces hierarchical clustering that exhibit a quality exceeding k-means and a hierarchical clustering algorithm. However, we identify several critical problem areas that would make it rather impractical to actually use such a hierarchy in a real-life environment. These predicaments point to the importance of semantic feature selection. Our main contribution is to test in details the applicability of ART to the important domain of hierarchical document clustering, an application of Adaptive Resonance that had received little attention until now.

show abstract

Section: Results With Independent Art1 Networkmentioning

confidence: 99%

Section: Results With Art1 Network In Seriesmentioning

confidence: 99%

Discovery of hierarchical thematic structure in text collections with adaptive resonance theory

Massey

2008

Neural Comput Applic

View full text Add to dashboard Cite

show abstract

“…Up to now many methods have been proposed in the field of text clustering including (Basu et al 2002;Buddeewong and Worapoj 2005;Jain et al 2004; Lee et al 2006;Shang et al 2006;Sun and Sun 2005;Wang and Zhang 2005;XU and Wang 2004). Some Text Classification methods also can be found in (Hung and Wermter 2003;Massey 2005;Song and Park 2006). Both text clustering and text classification include three phases as shown in Figure 1.…”

mentioning

confidence: 99%

A new unsupervised feature selection method for text clustering based on genetic algorithms

Shamsinejadbabki

Saraee

2011

J Intell Inf Syst

View full text Add to dashboard Cite

Nowadays a vast amount of textual information is collected and stored in various databases around the world, including the Internet as the largest database of all. This rapidly increasing growth of published text means that even the most avid reader cannot hope to keep up with all the reading in a field and consequently the nuggets of insight or new knowledge are at risk of languishing undiscovered in the literature. Text mining offers a solution to this problem by replacing or supplementing the human reader with automatic systems undeterred by the text explosion. It involves analyzing a large collection of documents to discover previously unknown information. Text clustering is one of the most important areas in text mining, which includes text preprocessing, dimension reduction by selecting some terms (features) and finally clustering using selected terms. Feature selection appears to be the most important step in the process. Conventional unsupervised feature selection methods define a measure of the discriminating power of terms to select proper terms from corpus.However up to now the valuation of terms in groups has not been investigated in reported works.In this paper a new and robust unsupervised feature selection approach is proposed that evaluates terms in groups. In addition a new Modified Term Variance measuring method is proposed for evaluating groups of terms. Furthermore a genetic based algorithm is designed and implemented for finding the most valuable groups of terms based on the new measure. These terms then will be utilized to generate the final feature vector for the clustering process . In order to evaluate and justify our approach the proposed method and also a conventional term variance method are implemented and tested using corpus collection Reuters-21578. For a more accurate comparison, methods have been tested on three corpuses and for each corpus clustering task has been done ten times and results are averaged. Results of comparing these two methods are very promising and show that our method produces better average accuracy and F1-measure than the conventional term variance method.

show abstract

“…ART converges to a stable representation after at most R−1 presentations of the R data items (Georgiopoulos et al 1990). However, an important and until now unresolved problem with ART stability was recently identified while investigating its application to a real-world problem (Massey 2005b). In short, the problem is that contrary to general belief, ART stability is not possible with infinite streaming data.…”

mentioning

confidence: 99%

“…Here, we present and analyse in detail the ART stabilisation problem we have previously identified briefly in Massey (2005b). Our contribution in this paper is to resolve this problem by presenting and testing a new stabilization strategy called conceptual duplication.…”

mentioning

confidence: 99%

Conceptual duplication

Massey

2007

Soft Comput

View full text Add to dashboard Cite

Stability and plasticity in learning systems are both equally essential, but achieving stability and plasticity simultaneously is difficult. Adaptive resonance theory (ART) neural networks are known for their plastic and stable learning of categories, hence providing an answer to the so called stability-plasticity dilemma. However, it has been demonstrated recently that contrary to general belief, ART stability is not possible with infinite streaming data. In this paper, we present an improved stabilization strategy for ART neural networks that does not suffer from this problem and that produces a soft-clustering solution as a positive side effect. Experimental results in a task of text clustering demonstrate that the new stabilization strategy works well, but with a slight loss in clustering quality compared to the traditional approach. For real-life intelligent applications in which infinite streaming data is generated, the stable and soft-clustering solution obtained with our approach more than outweighs the small loss in quality.

show abstract

Real-world text clustering with adaptive resonance theory neural networks

Cited by 13 publications

References 14 publications

Discovery of hierarchical thematic structure in text collections with adaptive resonance theory

Discovery of hierarchical thematic structure in text collections with adaptive resonance theory

A new unsupervised feature selection method for text clustering based on genetic algorithms

Conceptual duplication

Contact Info

Product

Resources

About