A fuzzy document clustering approach based on domain-specified ontology

Yue, Lin; Zuo, Wangmeng; Peng, Tao; Wang, Ying; Han, Xuming

doi:10.1016/j.datak.2015.04.008

Cited by 23 publications

(7 citation statements)

References 27 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, through the calculation and adjustment, we take c = 0.1, which can make the calculation results in the interval [0, 1] to disperse better.

r_{i j} = 1 ‐ c • d ((),, x_{i}, x_{j})

where d ( x i , x j ) is Euclidean distance,

d ((),, x_{i}, x_{j}) = \sqrt{true\sum {((), x_{i} - x_{j})}^{2} .}

Step 3Solving the transitive closure. The transitive closure of a fuzzy similarity relation R , also known as a fuzzy equivalence relation, is defined as the relation that is transitive, contains R , and has the smallest possible membership grades (Yue et al ., ). In this study, we solved the transitive closure using quadratic programming.…”

Section: Framework and Methodologymentioning

confidence: 99%

“…Step 3 Solving the transitive closure. The transitive closure of a fuzzy similarity relation R, also known as a fuzzy equivalence relation, is defined as the relation that is transitive, contains R, and has the smallest possible membership grades (Yue et al, 2015). In this study, we solved the transitive closure using quadratic programming.…”

Section: Main Step Of Fuzzy Clustering Analysis According Tomentioning

confidence: 99%

See 1 more Smart Citation

Risk prediction of product‐harm events using rough sets and multiple classifier fusion: an experimental study of listed companies in China

Wang

Zheng

et al. 2016

Expert Systems

View full text Add to dashboard Cite

With the increasing of frequency and destructiveness of product-harm events, study on enterprise crisis management becomes essentially important, but little literature thoroughly explores the risk-prediction method of product-harm event. In this study, an initial index system for risk prediction was built based on the analysis of the key drivers of the product-harm event's evolution; ultimately, nine risk-forecasting indexes were obtained using rough set attribute reduction. With the four indexes of cumulative abnormal returns as the input, fuzzy clustering was used to classify the risk level of a product-harm event into four grades. In order to control the uncertainty and instability of single classifiers in risk prediction, multiple classifiers fusion was introduced and combined with self-organising data mining (SODM). Further, an SODM-based multiple classifiers fusion (SB-MCF) model was presented for the risk prediction related to a product-harm event. The experimental results based on 165 Chinese listed companies indicated that the SB-MCF model improved the average predictive accuracy and reduced variation degree simultaneously. The statistical analysis demonstrated that the SB-MCF model significantly outperformed six widely used single classification models (e.g. neural networks, support vector machine, and case-based reasoning) and other six commonly used multiple classifiers fusion methods (e.g. majority voting, Bayesian method, and genetic algorithm).Keywords product-harm; risk prediction; multiple classifiers; self-organizing data mining; rough set IntroductionProduct-harm crises can be defined as 'discrete, well publicized occurrences wherein products are found to be defective or dangerous ' (Dawar & Pillutla, 2000). The increasing complexity of products, the increased demands of customers, and the greater vigilance on the part of the media have made product-harm crises increasingly visible occurrences (Klein & Dawar, 2004). In America, the frequency of product-harm events in the Although the probability of product-harm events is relatively low for an individual enterprise, such an event, if mishandled, will not only hurt the psychology and physiology of consumers but also lead to huge economic losses and a trust crisis for the enterprise. Further, it might have undesirable social and political impacts. For instance, in an investigation conducted by Sina.com, 84% of the interviewees stated that they would no longer buy Shuanghui meat products after the 'clenbuterol' scandal, while 7.8% of the interviewees were undecided. Only 8.4% of the interviewees stated that they would continue to buy Shuanghui products, which would lead to cumulative economic losses of over 20 billion yuan for the Shuanghui group. Similarly, because of the 'melamine scandal', Sanlu Group, China's third largest dairy company, went bankrupt. Moreover, this event led to public distrust and insecurity about food safety. According to a recent market report released by AC Nielsen about China's infant milk powder, China's brand o...

show abstract

“…In this paper, through the calculation and adjustment, we take c = 0.1, which can make the calculation results in the interval [0, 1] to disperse better.

r_{i j} = 1 ‐ c • d ((),, x_{i}, x_{j})

where d ( x i , x j ) is Euclidean distance,

d ((),, x_{i}, x_{j}) = \sqrt{true\sum {((), x_{i} - x_{j})}^{2} .}

Section: Framework and Methodologymentioning

confidence: 99%

Section: Main Step Of Fuzzy Clustering Analysis According Tomentioning

confidence: 99%

Risk prediction of product‐harm events using rough sets and multiple classifier fusion: an experimental study of listed companies in China

Wang

Zheng

et al. 2016

Expert Systems

View full text Add to dashboard Cite

show abstract

“…Hierarchical agglomerative clustering can be computationally expensive to use for large datasets. In [15] also proposed the use of semantic clustering to determine similar text documents. The text corpus is extracted and pre-processed.…”

Section: Literature Reviewmentioning

confidence: 99%

Semantics-based clustering approach for similar research area detection

et al. 2020

View full text Add to dashboard Cite

The manual process of searching out individuals in an already existing research field is cumbersome and time-consuming. Prominent and rookie researchers alike are predisposed to seek existing research publications in a research field of interest before coming up with a thesis. From extant literature, automated similar research area detection systems have been developed to solve this problem. However, most of them use keyword-matching techniques, which do not sufficiently capture the implicit semantics of keywords thereby leaving out some research articles. In this study, we propose the use of ontology-based pre-processing, Latent Semantic Indexing and K-Means Clustering to develop a prototype similar research area detection system, that can be used to determine similar research domain publications. Our proposed system solves the challenge of high dimensionality and data sparsity faced by the traditional document clustering technique. Our system is evaluated with randomly selected publications from faculties in Nigerian universities and results show that the integration of ontologies in preprocessing provides more accurate clustering results.

show abstract

“…As researchers find that the relations between words contribute to understanding the meaning of text, they construct a semantic network in terms of concepts, events, and their relations. Yue et al [15] constructed a domain-specific ontology to describe the hazards related to dairy products and translated the term-document vectors (namely, feature vectors of text) into a concept space. Wei et al [16] exploited an ontology hierarchical structure for word sense disambiguation to assess similarity of words.…”

Section: Feature Extractionmentioning

confidence: 99%

A Novel Text Clustering Approach Using Deep-Learning Vocabulary Network

Zhang

Zhao

et al. 2017

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Text clustering is an effective approach to collect and organize text documents into meaningful groups for mining valuable information on the Internet. However, there exist some issues to tackle such as feature extraction and data dimension reduction. To overcome these problems, we present a novel approach named deep-learning vocabulary network. The vocabulary network is constructed based on related-word set, which contains the “cooccurrence” relations of words or terms. We replace term frequency in feature vectors with the “importance” of words in terms of vocabulary network and PageRank, which can generate more precise feature vectors to represent the meaning of text clustering. Furthermore, sparse-group deep belief network is proposed to reduce the dimensionality of feature vectors, and we introduce coverage rate for similarity measure in Single-Pass clustering. To verify the effectiveness of our work, we compare the approach to the representative algorithms, and experimental results show that feature vectors in terms of deep-learning vocabulary network have better clustering performance.

show abstract

A fuzzy document clustering approach based on domain-specified ontology

Cited by 23 publications

References 27 publications

Risk prediction of product‐harm events using rough sets and multiple classifier fusion: an experimental study of listed companies in China

Risk prediction of product‐harm events using rough sets and multiple classifier fusion: an experimental study of listed companies in China

Semantics-based clustering approach for similar research area detection

A Novel Text Clustering Approach Using Deep-Learning Vocabulary Network

Contact Info

Product

Resources

About