Lihui Chen

Tri-state median filter for image denoising

In this work, a novel nonlinear filter, called tri-state median (TSM) filter, is proposed for preserving image details while effectively suppressing impulse noise. We incorporate the standard median (SM) filter and the center weighted median (CWM) filter into a noise detection framework to determine whether a pixel is corrupted, before applying filtering unconditionally. Extensive simulation results demonstrate that the proposed filter consistently outperforms other median filters by balancing the tradeoff between noise reduction and detail preservation.

Fuzzy clustering with weighted medoids for relational data

The well known k-medoids clustering approach groups objects through finding k representative objects based on the pairwise (dis)similarities of objects in the data set. In real applications, using only one object to capture or interpret each cluster may not be sufficient enough which in turn could affect the accuracy of the data analysis. In this paper, we propose a new fuzzy clustering approach called PFC for (dis)similarity-based data or relational data analysis. In PFC, objects in each fuzzy cluster carry various weights called prototype weights to represent their degrees of representativeness in that cluster. This mechanism enables each cluster to be represented by more than one objects. Compared with existing clustering approaches for relational data, PFC is able to capture the underlying structures of the data more accurately and provide richer information for the description of the resulting clusters. We introduce the detailed formulation of PFC and provide the analytical as well as experimental studies to demonstrate the merits of the proposed approach.

Watermark embedding in DC components of DCT for binary images

This paper investigates the feasibility of watermark embedding in the discrete cosine transform (DCT) domain for binary images. Watermark embedding is known to be difficult for binary images due to their binary nature. For frequency domain approach to binary image watermarking, a post-embedding binarization is a necessary step to ensure that the watermarked image is still a binary image. This step disturbs the watermark embedded and is likely to remove the watermark. We have succeeded in combating this interference by embedding watermarks in the DC components of DCT and employing a biased binarization threshold. This algorithm can be applied to binary images in general and experiments show that the embedding algorithm proposed can not only survive binarization, but also provide some degree of robustness against common image processing.

Clustering with Multiviewpoint-Based Similarity Measure

All clustering methods have to assume some cluster relationship among the data objects that they are applied on. Similarity between a pair of objects can be defined either explicitly or implicitly. In this paper, we introduce a novel multiviewpoint-based similarity measure and two related clustering methods. The major difference between a traditional dissimilarity/similarity measure and ours is that the former uses only a single viewpoint, which is the origin, while the latter utilizes many different viewpoints, which are objects assumed to not be in the same cluster with the two objects being measured. Using multiple viewpoints, more informative assessment of similarity could be achieved. Theoretical analysis and empirical study are conducted to support this claim. Two criterion functions for document clustering are proposed based on this new measure. We compare them with several well-known clustering algorithms that use other popular similarity measures on various document collections to verify the advantages of our proposal.

Dual Fuzzy-Possibilistic Co-clustering for Document Categorization

In this paper, we introduce a new algorithm called Dual Fuzzy-possibilistic Co-clustering (DFPC) for docu- ment categorization. The proposed algorithm offers several advantages. Firstly, the combined fuzzy and possibilistic cluster memberships in DFPC can provide realistic repre- sentation of document clusters. Secondly, as a co-clustering algorithm, DFPC can categorize high-dimensional datasets effectively. Thirdly, the possibilistic clustering element of the algorithm makes it robust to outliers. We detail the for- mulation of DFPC, and empirically demonstrate its effec- tiveness in categorizing benchmark document datasets.

Incremental Fuzzy Clustering With Multiple Medoids for Large Data

As an important technique of data analysis, clustering plays an important role in finding the underlying pattern structure embedded in unlabeled data. Clustering algorithms that need to store all the data into the memory for analysis become infeasible when the dataset is too large to be stored. To handle such large data, incremental clustering approaches are proposed. The key idea behind these approaches is to find representatives (centroids or medoids) to represent each cluster in each data chunk, which is a packet of the data, and final data analysis is carried out based on those identified representatives from all the chunks. In this paper, we propose a new incremental clustering approach called incremental multiple medoids-based fuzzy clustering (IMMFC) to handle complex patterns that are not compact and well separated. We would like to investigate whether IMMFC is a good alternative to capturing the underlying data structure more accurately. IMMFC not only facilitates the selection of multiple medoids for each cluster in a data chunk, but also has the mechanism to make use of relationships among those identified medoids as side information to help the final data clustering process. The detailed problem formulation, updating rules derivation, and the in-depth analysis of the proposed IMMFC are provided. Experimental studies on several large datasets that include real world malware datasets have been conducted. IMMFC outperforms existing incremental fuzzy clustering approaches in terms of clustering accuracy and robustness to the order of data. These results demonstrate the great potential of IMMFC for large-data analysis.

A Fuzzy Approach for Multitype Relational Data Clustering

Mining interrelated data among multiple types of objects or entities is important in many real-world applications. Despite extensive study on fuzzy clustering of vector space data, very limited exploration has been made on fuzzy clustering of relational data that involve several object types. In this paper, we propose a new fuzzy clustering approach for multitype relational data (FC-MR). In FC-MR, different types of objects are clustered simultaneously. An object is assigned a large membership with respect to a cluster if its related objects in this cluster have high rankings. In each cluster, an object tends to have a high ranking if its related objects have large memberships in this cluster. The FC-MR approach is formulated to deal with multitype relational data with various structures. The objective function of FC-MR is locally optimized by an efficient iterative algorithm, which updates the fuzzy membership matrix and the ranking matrix of one type at once while keeping those of other types constant. We also discuss the simplified FC-MR for multitype relational data with two special structures, namely, star-structure and extended star-structure. Experimental studies are conducted on benchmark document datasets to illustrate how the proposed approach can be applied flexibly under different scenarios in real-world applications. The experimental results demonstrate the feasibility and effectiveness of the new approach compared with existing ones.

Novelty detection for text documents using named entity recognition

In order to determine novel information from raw text documents, a novelty detection recommender system was developed to explore the method of comparing various types of entities within sentences. We first detected novel sentences using named entity recognition to extract the entity types of person, place, time, and organization. In addition, part-of-speech tagging was performed to tag each word in the documents, allowing syntactic structures of noun, verb, and adjective to be used for comparisons. WordNet, an English lexical database of concepts and relations, was also incorporated to generate synonyms for the entities and parts of speech as well as to determine the similarity of sentences. The novelty score of each sentence was determined by using two different metrics, UniqueComparison and Importance Value. UniqueComparison calculated the number of matched entities, whereas ImportanceValue took into account the total weight of matched words that coexisted in both the test and history sentences. The results look promising when compared to the benchmark scores for the Text Retrieval Conferences (TREC) Novelty Track 2004. This demonstrated that the combination of named entity recognition and part-of-speech tagging is capable of detecting novelty with good results.

Possibilistic fuzzy co-clustering of large document collections

In this paper we propose a new co-clustering algorithm called possibilistic fuzzy co-clustering (PFCC) for automatic categorization of large document collections. PFCC integrates a possibilistic document clustering technique and a combined formulation of fuzzy word ranking and partitioning into a fast iterative co-clustering procedure. This novel framework brings about simultaneously some benefits including robustness in the presence of document and word outliers, rich representations of co-clusters, highly descriptive document clusters, a good performance in a high-dimensional space, and a reduced sensitivity to the initialization in the possibilistic clustering. We present the detailed formulation of PFCC together with the explanations of the motivations behind. The advantages over other existing works and the algorithms proof of convergence are provided. Experiments on several large document data sets demonstrate the effectiveness of PFCC.

A heuristic-based fuzzy co-clustering algorithm for categorization of high-dimensional data

Fuzzy co-clustering is a technique that performs simultaneous fuzzy clustering of objects and features. It is known to be suitable for categorizing high-dimensional data, due to its dynamic dimensionality reduction mechanism achieved through simultaneous feature clustering. We introduce a new fuzzy co-clustering algorithm called Heuristic Fuzzy Co-clustering with the Ruspinis condition (HFCR), which addresses several issues in some prominent existing fuzzy co-clustering algorithms. Among these issues are the performance on data sets with overlapping feature clusters and the unnatural representation of feature clusters. The key idea behind HFCR is the formulation of the dual-partitioning approach for fuzzy co-clustering, replacing the existing partitioning-ranking approach. HFCR adopts an efficient and practical heuristic method that can be shown to be more robust than our earlier effort for the dual-partitioning approach. We explain the proposed algorithm in details and provide an analytical study on its advantages. Experimental results on 10 large benchmark document data sets confirm the effectiveness of the new algorithm.