Most recent issue published online for the International Journal of Knowledge Engineering and Soft Data Paradigms.

Extraction of the contents in the web texts by content-density distribution

Saori Kitahara; Koya Tamura; Kenji Hatano — 2012-03-01T23:20:50-05:00

Extraction of the contents in the web texts by content-density distribution
Saori Kitahara; Koya Tamura; Kenji Hatano
International Journal of Knowledge Engineering and Soft Data Paradigms, Vol. 3, No. 2 (2011) pp. 108 - 120
In recent years, users use result snippets of a web search engine to grasp the content of web pages, when users search for useful information on the internet. However, they are sometimes unable to notice the content of web pages by reading the result snippets because these snippets are so short that they cannot determine whether the content of each web page is relevant. To address this problem, we propose a method for grasping the content of each web page and extracting a part of the web page concerned to query keywords. This method is more effective than conventional methods based on snippets, because we regard the content as a set of words in the text of a web page, and we generate the content-density distribution by using both the position and the influence of the word. In the result of our experiments, we found that our method is useful for gasping the influence of extracted web text.

Hierarchical clustering algorithm with combined criteria for large and complex similarity data

Kensuke Tanioka; Hiroshi Yadohisa — 2012-03-01T23:20:50-05:00

Hierarchical clustering algorithm with combined criteria for large and complex similarity data
Kensuke Tanioka; Hiroshi Yadohisa
International Journal of Knowledge Engineering and Soft Data Paradigms, Vol. 3, No. 2 (2011) pp. 121 - 131
Recent developments in information technology have enabled us to deal with large and complex similarity data. Researchers often need to know the cluster structures of such a datasets before constructing inferential models or other such interrogation techniques. To reveal cluster structures, Chamelleon (Karypis et al., 1999) can make subclusters from datasets using graph partition methods and apply hierarchical clustering to reduce the amount of calculations and to reflect the structures in the clusters. Chamelleon can capture arbitrary shaped clusters from similarity data. It can consider intrasimilarities and intersimilarities when two clusters are combined. In addition, the method is robust for outliers whose objects are far from other objects in the same subcluster. However, it cannot detect the cluster structures that cannot be detected by the group average method. This paper proposes a new hierarchical clustering method based on the single linkage method for use when similarity data and subclusters are given. The proposed method has three advantages. First, it considers the intrasimilarities and intersimilarities of some parts in subclusters. Second, it considers the effects of outliers and cluster sizes. Finally, it detects arbitrary shaped cluster structures that cannot be detected by Chamelleon.

Dissimilarity criteria in hierarchical clustering for interval-valued functional data

Nobuo Shimizu — 2012-03-01T23:20:50-05:00

Dissimilarity criteria in hierarchical clustering for interval-valued functional data
Nobuo Shimizu
International Journal of Knowledge Engineering and Soft Data Paradigms, Vol. 3, No. 2 (2011) pp. 132 - 142
We deal with hierarchical clustering for interval-valued functional data. Functional data is defined as the data which is function, or as the data approximated as a function. Functional clustering is proposed as clustering for functional data. Interval-valued functional data is defined as the functional data whose range corresponding to each value in the domain is interval-valued data. Interval-valued data is especially typical in symbolic data, and also intervalvalued functional data can be considered to be a kind of symbolic data. We propose some new dissimilarity criteria in hierarchical clustering for intervalvalued functional data as the extension of functional clustering method, and apply these criteria to real data.

Transform of visual analogue scale data and their clustering

2012-03-01T23:20:50-05:00

Transform of visual analogue scale data and their clustering
Kotoe Katayama; Rui Yamaguchi; Seiya Imoto; Keiko Matsuura; Kenji Watanabe; Satoru Miyano
International Journal of Knowledge Engineering and Soft Data Paradigms, Vol. 3, No. 2 (2011) pp. 143 - 151
We propose a hierarchical clustering for the visual analogue scale (VAS) in the framework of symbolic data analysis (SDA). The VAS is a method that can be readily understood by most people to measure a characteristic or attitude that cannot be directly measured. VAS is of most value when looking at change within the same people, and is of less value for comparing across a group of people because they have different sense. It could be argued that a VAS is trying to produce interval/ratio data out of subjective values that are at best ordinal. Thus, some caution is required in handling VAS. We describe VAS as distribution and handle it as new type data in SDA. SDA was proposed by Diday at the end of the 1980s and is a new approach for analysing huge and complex data. In SDA, an observation is described by not only numerical values but also 'higher-level units'; sets, intervals, distributions, etc. In this paper, we define 'VAS distribution' and 'VAS changes distribution' as new type data in SDA and propose a hierarchical clustering for these new type data.

Implementation of various data processing and evaluation techniques on ICDDR,B surveillance data to generate optimal decision tree for patients classification

Rashedur M. Rahman; Fazle Rabbi Md. Hasan — 2012-03-01T23:20:50-05:00

Implementation of various data processing and evaluation techniques on ICDDR,B surveillance data to generate optimal decision tree for patients classification
Rashedur M. Rahman; Fazle Rabbi Md. Hasan
International Journal of Knowledge Engineering and Soft Data Paradigms, Vol. 3, No. 2 (2011) pp. 152 - 187
The surveillance system established in ICDDR,B collect information of diarrhoeal disease. Our research focuses on generating decision tree models to categorise diarrhoeal patients according to the severance of disease. From the decision tree generated based on earlier cases stored in the surveillance data, decision rules are generated. These rules are used to classify patients into three classes according to their criticality: high, mid, low. This would help the hospital authority to take prudent actions on critical patients. Different techniques are used to build an optimal decision tree by considering various set of data, and generated trees are compared with various performance metrics, e.g., accuracy, precision, recall, area under ROC curve, etc.

Alternative fuzzy c-lines and local principal component extraction

Katsuhiro Honda; Sakuya Nakao; Akira Notsu; Hidetomo Ichihashi — 2012-03-01T23:20:50-05:00

Alternative fuzzy c-lines and local principal component extraction
Katsuhiro Honda; Sakuya Nakao; Akira Notsu; Hidetomo Ichihashi
International Journal of Knowledge Engineering and Soft Data Paradigms, Vol. 3, No. 2 (2011) pp. 188 - 200
Alternative c-means is an extension of k-means-type clustering for robustifying cluster estimation, in which a modified distance measure instead of the conventional Euclidean distance is used based on the robust M-estimation concept. In this paper, alternative c-means is further extended to linear clustering models with line-shape prototypes, in which the clustering criteria of distances between data samples and linear prototypes are calculated by the lower rank approximation concept. The iterative updating scheme is derived in a pseudo-M-estimation procedure with a weight function for the modified distance measure and is demonstrated to be useful for extracting linear substructures from noisy datasets. In numerical experiments, the model is applied to POS transaction data analysis based on local PCA-like data summarisation.