Search

This paper presents a web application that seeks to facilitate the selection of research articles that are relevant or not to a topic. The process starts when a researcher writes a search string, which is sent to the Scopus API. With the results obtained, a grouping process is carried out to generate a visualization by groups or topics instead of the traditional ordered lists of results, making it easier for users to discard groups of articles irrelevant to their query. The proposal uses five clustering algorithms, among which Spectral and K-means exhibited the best performance in classical information retrieval metrics on four state of the art datasets. The application was assessed in two rounds by researchers of Universidad del Cauca, who, in the final round, considered that 71.4% of the clusters had a good title, 92.9% of the clusters had a good document order, and 65.8% of the articles were well clustered. The implementation of overlapping in grouping stands out since it allows articles to belong to several topics. Finally, the results are promising, and the application constitutes a valuable contribution for researchers in developing their projects. However, the results are not generalizable, and the need to create better labeling algorithms to generate more descriptive titles is evident, along with the use of tools to assist the user in query construction. [ABSTRACT FROM AUTHOR]

Document Indexing is but not limited to summarizing document contents with a small set of keywords or concepts of a knowledge base. Such a compact representation of document contents eases their use in numerous processes such as content-based information retrieval, corpus-mining and classification. An important effort has been devoted in recent years to (partly) automate semantic indexing, i.e. associating concepts to documents, leading to the availability of large corpora of semantically indexed documents. In this paper we introduce a method that hierarchically clusters documents based on their semantic indices while providing the proposed clusters with semantic labels. Our approach follows a neighbor joining strategy. Starting from a distance matrix reflecting the semantic similarity of documents, it iteratively selects the two closest clusters to merge them in a larger one. The similarity matrix is then updated. This is usually done by combining similarity of the two merged clusters, e.g. using the average similarity. We propose in this paper an alternative approach where the new cluster is first semantically annotated and the similarity matrix is then updated using the semantic similarity of this new annotation with those of the remaining clusters. The hierarchical clustering so obtained is a binary tree with branch lengths that convey semantic distances of clusters. It is then post-processed by using the branch lengths to keep only the most relevant clusters. Such a tool has numerous practical applications as it automates the organization of documents in meaningful clusters (e.g. papers indexed by MeSH terms, bookmarks or pictures indexed by WordNet) which is a tedious everyday task for many people. We assess the quality of the proposed methods using a specific benchmark of annotated clusters of bookmarks that were built manually. Each dataset of this benchmark has been clustered independently by several users. Remarkably, the clusters automatically built by our method are congruent with the clusters proposed by experts. All resources of this work, including source code, jar file, benchmark files and results are available at this address: http://sc.nicolasfiorini.info . [ABSTRACT FROM AUTHOR]

I present a fast algorithm for modeling invasion percolation (IP) with trapping (TIP). IP is a numerical algorithm that models quasi-static (i.e. slow) fluid invasion in porous media. Trapping occurs when the invading fluid (that is injected) forms continuous surfaces surrounding patches of the displaced fluid (that is assumed incompressible and originally saturates the invaded medium). In TIP, the invading fluid is not allowed to enter the trapped patches. I demonstrate that TIP can be modeled in two steps: (1) Run an IP simulation without trapping (NTIP). (2) Identify the sites that invaded trapped regions and remove them from the chronological list of sites invaded in NTIP. Fast algorithms exist for solving NTIP. The focus of this paper is to propose an efficient solution for step (2). I show that it can be solved using a disjoint set data structure and going backward in time, i.e. by un-invading all sites invaded in NTIP in reverse order. Time reversal of the invasion greatly reduces the computational complexity for the identification of trapped sites as one only needs to investigate sites neighbor to the latest invaded/un-invaded site. This differs from traditional approaches where trapping is performed in real time, i.e. as the IP simulation is running, and where it is sometimes necessary to investigate the whole lattice to identify newly trapped regions. With the proposed algorithm, the total computational time for the identification and the removal of trapped sites goes as O ( N ), where N is the total number of sites in the lattice. [ABSTRACT FROM AUTHOR]

Document clustering refers to grouping similar documents together automatically. Labels of the clusters, usually edited manually, are helpful for users to quickly grasp the major meaning of the grouped documents. Therefore, high quality labels are desired in many user-facing applications. However, assigning the labels manually is time consuming and tedious. In this paper a hybrid approach is proposed to automate the labeling process. First, linguistic knowledge are used to ensure candidate labels’ readability and information quantity by exploring the dependencies between words. Second, a statistical generative model is proposed to select representative labels. It scores a label w.r.t. a cluster by estimating how likely the cluster is generated by the label. The proposed approach is evaluated on two data sets in both English and Chinese. Experimental results show that the proposed approach produces high quality labels and outperforms existing state-of-art methods on both manual and automatic evaluations. [ABSTRACT FROM AUTHOR]

As an important boundary-based clustering algorithm, support vector clustering (SVC) can benefit many real applications owing to its capability of handling arbitrary cluster shapes, especially those directly or indirectly related to pattern exploration and description. As the application deepens, the importance of performance (i.e. criterions of accuracy and efficiency) of SVC increases. To identify gaps in the current methods and propose novel research directions for SVC, we present a survey of the literature in this area. Our approach is to classify the most recent advances into either theory or application. For theoretical contributions, advances related to parameter selection and optimization, dual-problem solutions, and cluster labeling are introduced. We also simultaneously summarize the advantages and drawbacks of each study. With respect to applications, we clearly describe eight groups of schemes based on SVC, either as individual or hybrid methods. Finally, we identify the gaps in SVC research and suggest several future research issues and trends. [ABSTRACT FROM AUTHOR]

Abstract: Efficient clustering algorithms have been developed to automatically group documents into subgroups (clusters). Once clustering has been performed, an important additional step is to help users make sense of the obtained clusters. Existing methods address this issue by assigning to each cluster a flat list of descriptive terms (labels) that are extracted from the documents, most often using statistical techniques borrowed from the field of feature selection or reduction. A limitation of these unstructured descriptions of clusters’ contents is that they do not account for the meaningful relationships between the terms. In contrast, we propose a graph representation, which makes the clusters easier to interpret by putting the descriptive terms in context, and by performing some simple network analysis. Our experiments reveal that the proposed method allows for a deeper level of interpretation, both when the clusters under study are homogeneous and when they are heterogeneous. In addition, evaluation procedures presented in the paper show that the graph-based representation of each cluster, while being very synthetic, still quite faithfully reflects the original content of the cluster. [Copyright &y& Elsevier]

Abstract: The Web is overcrowded with news articles, an overwhelming information source both with its amount and diversity. Document clustering is a powerful technique that has been widely used for organizing data into smaller and manageable information kernels. Several approaches have been proposed which, however, suffer from problems like synonymy, ambiguity and lack of a descriptive content marking of the generated clusters. In this work, we are investigating the application of a great spectrum of clustering algorithms, as well as similarity measures, to news articles that originate from the Web. Also, we are proposing the enhancement of standard k-means algorithm using the external knowledge from WordNet hypernyms in a twofold manner: enriching the “bag of words” used prior to the clustering process and assisting the label generation procedure following it. Furthermore, we are examining the effect that text preprocessing has on clustering. Operating on a corpus of news articles derived from major news portals, our comparison of the existing clustering methodologies revealed that k-means, gives better aggregate results when it comes to efficiency. This is amplified when the algorithm is accompanied with preliminary steps for data cleaning and normalizing, despite its simple nature. Moreover, the proposed WordNet-enabled W-k means clustering algorithm significantly improves standard k-means generating also useful and high quality cluster tags by using the presented cluster labeling process. [Copyright &y& Elsevier]

Abstract: Research on the extraction of content relations from text corpora is a high-priority topic in natural language processing. This is not surprising since content relations form the backbone of any ontology, and ontologies are increasingly made use of in knowledge-based applications. However, so far most of the works focus on the detection of a restricted number of prominent verbal relations, including in particular is-a, has-part and cause. Our application, which aims to provide comprehensive, easy-to-understand content representations of complex functional objects described in patent claims, faces the need to derive a large number of content relations that cannot be limited a priori. To cope with this problem, we take advantage of the fact that deep syntactic dependency structures of sentences capture all relevant content relations—although without any abstraction. We implement thus a three-step strategy. First, we parse the claims to retrieve the deep syntactic dependency structures from which we then derive the content relations. Second, we generalize the obtained relations by clustering them according to semantic criteria, with the goal to unite all sufficiently similar relations. Finally, we identify a suitable name for each generalized relation. To keep the scope of the article within reasonable limits and to allow for a comparison with state-of-the-art techniques, we focus on verbal relations. [Copyright &y& Elsevier]

Abstract: In many support vector-based clustering algorithms, a key computational bottleneck is the cluster labeling time of each data point which restricts the scalability of the method. In this paper, we review a general framework of support vector-based clustering using dynamical system and propose a novel method to speed up labeling time which is log-linear to the size of data. We also give theoretical background of the proposed method. Various large-scale benchmark results are provided to show the effectiveness and efficiency of the proposed method. [Copyright &y& Elsevier]

In this work, we address the issue of reconfigurability of a hand-gesture recognition system. The calibration or setup of the operational parameters of such a system is a time-consuming effort, usually performed by trial and error, and often causing system performance to suffer because of designer impatience. In this work, we suggest a methodology using a neighborhood-search algorithm for tuning system parameters. Thus, the design of hand-gesture recognition systems is transformed into an optimization problem. To test the methodology, we address the difficult problem of simultaneous calibration of the parameters of the image processing/fuzzy C-means (FCM) components of a hand-gesture recognition system. In addition, we proffer a method for supervising the FCM algorithm using linear programming and heuristic labeling. Resulting solutions exhibited fast convergence (in the order of ten iterations) to reach recognition accuracies within several percent of the optimal. Comparative performance testing using three gesture databases (BGU, American Sign Language and Gripsee), and a real-time implementation (Tele-Gest) are reported on. [ABSTRACT FROM AUTHOR]

We extend the Hoshen–Kopelman (HK) algorithm for cluster labeling to non-lattice environments, where sites are placed at random at non-lattice points. This extension is useful for continuum systems and disordered networks. Our extension of the HK algorithm relies on several data structures that describe network connectivity regardless of its dimensionality. Just as for the classic HK algorithm on lattices, our extension is completed in a single pass through the sites of the network and cluster relabeling operates on a vector whose size is much smaller than the size of the network. Our extension of the HK algorithm works for any environment (lattice or non-lattice) of any dimensionality, type (sites, bonds or both), and with arbitrary connectivity between the sites. The proposed extension is illustrated through a simple network consisting of 16 sites and 24 bonds, and applied to a complex network extracted from a 3D micro-focused X-ray CT image of Bentheimer sandstone consisting of 3677 sites and 8952 bonds. [Copyright &y& Elsevier]

Abstract: In this paper, two examples are given where the cluster labeling algorithm of Al-Futaisi and Patzek for arbitrary networks produces incorrect results. The reason for this is pointed out, and an improvement to the algorithm is proposed so that such errors can be avoided in future. [Copyright &y& Elsevier]