Categorical Clustering Research Articles

Urdu ligature recognition using multi-level agglomerative hierarchical clustering

Optical character recognition (OCR) system holds great significance in human-machine interaction. OCR has been the subject of intensive research especially for Latin, Chinese and Japanese script. Comparatively, little work has been done for Urdu OCR, due to the complexities and segmentation errors associated with its cursive script. This paper proposes an Urdu OCR system which aims at ligature-level recognition of Urdu text. This ligature based recognition approach overcomes the character-levelsegmentation problems associated with cursive scripts. A newly developed OCR algorithm is introduced that uses a semi-supervised multi-level clustering for categorization of the ligatures. Classification is performed using four machine learning techniques i.e. decision trees, linear discriminant analysis, naive Bayes and k-nearest neighbor (K-NN). The system was implemented and the results show 62, 61, 73 and 90% accuracy for decision tree, linear discriminant analysis, naive Bayes and K-NN respectively.

Mapping forest fires by nonparametric clustering analysis

Fires have a noteworthy role to play with regards to ecological and environmental losses in Mediterranean forests. In addition to ecological impacts, fire may create economic, social as well as cultural changes. The detection of fire-scars has critical importance to help decrease losses. In the present study, forest fires recorded in Antalya, one of the most important ecological and tourist regions within the Western Mediterranean, were clustered and mapped. Since the dominant factors and devastation records derived from the cases had nominal-scaled properties, a categorical data-based nonparametric clustering algorithm was performed in this evaluation. The proposed tool, k-modes algorithm, uses modes instead of means for clustering. The algorithm may be implemented quickly and does not make distributional assumptions concerning the available data. It uses a frequency-based method to update the modes of the fires. The derived modes from the maps may be useful information for local authorities to manage. In conclusion, the proposed nonparametric clustering procedure may be employed to build a decision-support system to monitor and identify fire activities and to enhance fire management efficiency.

Order or chaos? Understanding career mobility using categorical clustering and information theory

We examine the careers of a nationally representative US cohort of young adults using sequence analysis and information-theoretic techniques to describe these careers’ structure and how this structure might inform differences in wage mobility. We operationalise the career as a sequence of industry-occupation pairs observed quarterly. We investigate how the content of these pairs and their organisation over time relate to future mobility. We perform the analysis across three different mobility groups, one of which is characterised by persistent low-wage work. Contrary to what one might expect, low-wage work is not typified by a lack of structure, even in many of the careers in which the worker is weakly attached to the labour market. Using clustering techniques customised to this problem, we build a typology of careers within three groups of workers defined by their wage mobility. We find significant variation within, as well as similarity across the three groups, enhancing our understanding of careers with different levels of mobility.

A Fast and Efficient Method for Training Categorical Radial Basis Function Networks

This brief presents a novel learning scheme for categorical data based on radial basis function (RBF) networks. The proposed approach replaces the numerical vectors known as RBF centers with categorical tuple centers, and employs specially designed measures for calculating the distance between the center and the input tuples. Furthermore, a fast noniterative categorical clustering algorithm is proposed to accomplish the first stage of RBF training involving categorical center selection, whereas the weights are calculated through linear regression. The method is applied on 22 categorical data sets and compared with several different learning schemes, including neural networks, support vector machines, naïve Bayes classifier, and decision trees. Results show that the proposed method is very competitive, outperforming its rivals in terms of predictive capabilities in the majority of the tested cases.

Erratum to "Categorical time series clustering: Case study of Korean pro-baseball data"

Erratum to "Categorical time series clustering: Case study of Korean pro-baseball data"

Neural Categorization of Vibrotactile Frequency in Flutter and Vibration Stimulations: An fMRI Study.

As the use of wearable haptic devices with vibrating alert features is commonplace, an understanding of the perceptual categorization of vibrotactile frequencies has become important. This understanding can be substantially enhanced by unveiling how neural activity represents vibrotactile frequency information. Using functional magnetic resonance imaging (fMRI), this study investigated categorical clustering patterns of the frequency-dependent neural activity evoked by vibrotactile stimuli with gradually changing frequencies from 20 to 200 Hz. First, a searchlight multi-voxel pattern analysis (MVPA) was used to find brain regions exhibiting neural activities associated with frequency information. We found that the contralateral postcentral gyrus (S1) and the supramarginal gyrus (SMG) carried frequency-dependent information. Next, we applied multidimensional scaling (MDS) to find low-dimensional neural representations of different frequencies obtained from the multi-voxel activity patterns within these regions. The clustering analysis on the MDS results showed that neural activity patterns of 20-100 Hz and 120-200 Hz were divided into two distinct groups. Interestingly, this neural grouping conformed to the perceptual frequency categories found in the previous behavioral studies. Our findings therefore suggest that neural activity patterns in the somatosensory cortical regions may provide a neural basis for the perceptual categorization of vibrotactile frequency.

범주형 시계열 자료의 군집화: 프로야구 자료의 사례 연구

범주형 시계열 자료의 군집화: 프로야구 자료의 사례 연구

Revealing structures in narratives: A mixed-methods approach to studying interdisciplinary handoff in critical care

ObjectiveTo examine a novel mixed-methods approach for studying patterns of clinical communication that could inform future informatics solutions, with a specific focus on handoff within interdisciplinary teams. Materials and methodsResearchers observed, recorded, and transcribed verbal handoff discussions of different members of critical care teams. The transcripts were coded qualitatively, and then analyzed quantitatively for emerging structural patterns using categorical cluster analysis, and for degree of shared mental models (SMM) using the modified Pyramid method. ResultsAn empirical study using the proposed mixed-methods approach suggested emerging patterns of communication among clinicians. For example, the temporal focus of handoff was often determined by the role of the clinician giving the handoff; the clinical content of handoff was consistent between clinicians, but varied between patients. The SMM index ranged from 0.065 (with the maximum possible overlap score of 1) to 0.007 with a median of 0.026; the overlap was higher in statements concerned with patient presentation (23.6% of these had overlap) and referring to the past (24% overlapped). This calculated SMM index was correlated with the assessment of coherence within the participating teams by independent physicians (r=0.63, p=0.038). ConclusionsThe proposed novel mixed-methods approach helped to reveal emerging patterns in content and structure of handoff communication and highlight differences due to the clinical context, and to the different priorities of clinicians on interdisciplinary patient care teams. The approach for calculating SMM is more ecologically sensitive as it relies on naturally occurring discourse and less intrusive than traditional ways of assessing SMM, and takes initial steps toward establishing empirical foundation for the design of electronic tools to support handoff in interdisciplinary teams.

Cluster structures on strata of flag varieties

We introduce some new Frobenius subcategories of the module category of a preprojective algebra of Dynkin type, and we show that they have a cluster structure in the sense of Buan–Iyama–Reiten–Scott. These categorical cluster structures yield cluster algebra structures in the coordinate rings of intersections of opposite Schubert cells.

Evaluation of Modified Categorical Data Fuzzy Clustering Algorithm on the Wisconsin Breast Cancer Dataset

The early diagnosis of breast cancer is an important step in a fight against the disease. Machine learning techniques have shown promise in improving our understanding of the disease. As medical datasets consist of data points which cannot be precisely assigned to a class, fuzzy methods have been useful for studying of these datasets. Sometimes breast cancer datasets are described by categorical features. Many fuzzy clustering algorithms have been developed for categorical datasets. However, in most of these methods Hamming distance is used to define the distance between the two categorical feature values. In this paper, we use a probabilistic distance measure for the distance computation among a pair of categorical feature values. Experiments demonstrate that the distance measure performs better than Hamming distance for Wisconsin breast cancer data.

Categories in the pigeon brain: A reverse engineering approach.

Pigeons are well known for their visual capabilities as well as their ability to categorize visual stimuli at both the basic and superordinate level. We adopt a reverse engineering approach to study categorization learning: Instead of training pigeons on predefined categories, we simply present stimuli and analyze neural output in search of categorical clustering on a solely neural level. We presented artificial stimuli, pictorial and grating stimuli, to pigeons without the need of any differential behavioral responding while recording from the nidopallium frontolaterale (NFL), a higher visual area in the avian brain. The pictorial stimuli differed in color and shape; the gratings differed in spatial frequency and amplitude. We computed representational dissimilarity matrices to reveal categorical clustering based on both neural data and pecking behavior. Based on neural output of the NFL, pictorial and grating stimuli were differentially represented in the brain. Pecking behavior showed a similar pattern, but to a lesser extent. A further subclustering within pictorial stimuli according to color and shape, and within gratings according to frequency and amplitude, was not present. Our study gives proof-of-concept that this reverse engineering approach-namely reading out categorical information from neural data--can be quite helpful in understanding the neural underpinnings of categorization learning.

Soft subspace clustering of categorical data with probabilistic distance

Categorical data clustering is an important subject in pattern recognition. Currently, subspace clustering of categorical data remains an open problem due to the difficulties in estimating attribute interestingness according to the statistics of categories in clusters. In this paper, a new algorithm is proposed for clustering categorical data with a novel soft feature-selection scheme, by which each categorical attribute is automatically assigned a weight that correlates with the smoothed dispersion of the categories in a cluster. In the proposed algorithm, dissimilarity between categorical data objects is measured using a probabilistic distance function, based on kernel density estimation for categorical attributes. We also make use of the probabilistic distances to define a cluster validity index for estimating the number of categorical clusters. The suitability of the proposal is demonstrated in an empirical study done with some widely used real-world data sets and synthetic data sets, and the results show its outstanding performance.

Space Structure and Clustering of Categorical Data.

Learning from categorical data plays a fundamental role in such areas as pattern recognition, machine learning, data mining, and knowledge discovery. To effectively discover the group structure inherent in a set of categorical objects, many categorical clustering algorithms have been developed in the literature, among which k -modes-type algorithms are very representative because of their good performance. Nevertheless, there is still much room for improving their clustering performance in comparison with the clustering algorithms for the numeric data. This may arise from the fact that the categorical data lack a clear space structure as that of the numeric data. To address this issue, we propose, in this paper, a novel data-representation scheme for the categorical data, which maps a set of categorical objects into a Euclidean space. Based on the data-representation scheme, a general framework for space structure based categorical clustering algorithms (SBC) is designed. This framework together with the applications of two kinds of dissimilarities leads two versions of the SBC-type algorithms. To verify the performance of the SBC-type algorithms, we employ as references four representative algorithms of the k -modes-type algorithms. Experiments show that the proposed SBC-type algorithms significantly outperform the k -modes-type algorithms.

Improving the Decision Value of Hierarchical Text Clustering Using Term Overlap Detection

Humans are used to expressing themselves with written language and language provides a medium with which we can describe our experiences in detail incorporating individuality. Even though documents provide a rich source of information, it becomes very difficult to identify, extract, summarize and search when vast amounts of documents are collected especially over time. Document clustering is a technique that has been widely used to group documents based on similarity of content represented by the words used. Once key groups are identified further drill down into sub-groupings is facilitated by the use of hierarchical clustering. Clustering and hierarchical clustering are very useful when applied to numerical and categorical data and cluster accuracy and purity measures exist to evaluate the outcomes of a clustering exercise. Although the same measures have been applied to text clustering, text clusters are based on words or terms which can be repeated across documents associated with different topics. Therefore text data cannot be considered as a direct ‘coding’ of a particular experience or situation in contrast to numerical and categorical data and term overlap is a very common characteristic in text clustering. In this paper we propose a new technique and methodology for term overlap capture from text documents, highlighting the different situations such overlap could signify and discuss why such understanding is important for obtaining value from text clustering. Experiments were conducted using a widely used text document collection where the proposed methodology allowed exploring the term diversity for a given document collection and obtain clusters with minimum term overlap.

Coupled attribute similarity learning on categorical data.

Attribute independence has been taken as a major assumption in the limited research that has been conducted on similarity analysis for categorical data, especially unsupervised learning. However, in real-world data sources, attributes are more or less associated with each other in terms of certain coupling relationships. Accordingly, recent works on attribute dependency aggregation have introduced the co-occurrence of attribute values to explore attribute coupling, but they only present a local picture in analyzing categorical data similarity. This is inadequate for deep analysis, and the computational complexity grows exponentially when the data scale increases. This paper proposes an efficient data-driven similarity learning approach that generates a coupled attribute similarity measure for nominal objects with attribute couplings to capture a global picture of attribute similarity. It involves the frequency-based intra-coupled similarity within an attribute and the inter-coupled similarity upon value co-occurrences between attributes, as well as their integration on the object level. In particular, four measures are designed for the inter-coupled similarity to calculate the similarity between two categorical values by considering their relationships with other attributes in terms of power set, universal set, joint set, and intersection set. The theoretical analysis reveals the equivalent accuracy and superior efficiency of the measure based on the intersection set, particularly for large-scale data sets. Intensive experiments of data structure and clustering algorithms incorporating the coupled dissimilarity metric achieve a significant performance improvement on state-of-the-art measures and algorithms on 13 UCI data sets, which is confirmed by the statistical analysis. The experiment results show that the proposed coupled attribute similarity is generic, and can effectively and efficiently capture the intrinsic and global interactions within and between attributes for especially large-scale categorical data sets. In addition, two new coupled categorical clustering algorithms, i.e., CROCK and CLIMBO are proposed, and they both outperform the original ones in terms of clustering quality on UCI data sets and bibliographic data.

Evaluation of selected approaches to clustering categorical variables

This paper focuses on recently proposed similarity measures and their performance in categorical variable clustering. It compares clustering results using three recently developed similarity measures (IOF, OF and Lin measures) with results obtained using two association measures for nominal variables (Cramér’s V and the uncertainty coefficient) and with the simple matching coefficient (the overlap measure). To eliminate the influence of a particular linkage method on the structure of final clusters, three linkage methods are examined (complete, single, average). The created groups (clusters) of variables can be considered as the basis for dimensionality reduction, e.g. by choosing one of the variables from a given group as a representative for the whole group. The quality of resulting clusters is evaluated by the within-cluster variability, expressed by the WCM coefficient, and by dendrogram analysis. The examined similarity measures are compared and evaluated using two real data sets from a social survey.

Adaptation of Periurban Cattle Production Systems to Environmental Changes: Feeding Strategies of Herdsmen in Southern Benin

We undertook a survey of 112 cattle farms in the periphery of Cotonou in Benin, with the aim to characterize the diversity of the production systems and to better understand herders’ feeding strategies and perspectives. We used the categorical component analysis and two-step clustering techniques to classify the farms into homogeneous groups. The adaptive strategy of a given farm was predicted through logistic regression analysis technique. Four distinct farm types were identified as follows: large integrated agro-silvopastoral (17%), small agro-silvopastoral (28%), pastoral (30%), and silvopastoral (25%). Irrespective of farm type, low availability, and difficult access to pasture due to increased crop/vegetable farming, climate variability and urbanization were commonly reported by all herders. Current coping strategies included the use of lowlands pastoral resources and exploring new grazing routes. Future strategies included moving animals towards rural locations and significantly (p < 0.001) depended on farm type and its proximity to urban centers.

Art Video Games

This study conducted analytical and semiexperimental research with the purpose of testing if art video games serve as a form of transmission of social representations and feelings. Accordingly, a free-association questionnaire was used after participants played the game The Graveyard. The associative method was paired with item hierarchization and clustering techniques using a structural approach. The data were analyzed using mixed methods (frequency analysis, semantic weight, and categorical clustering). Additionally, a cluster analysis was conducted to determine connections between representations. Afterward, categories were compared with the representations the game designers wanted to transmit to the players. The results of the research confirmed the possibility of accessing people’s social representations using a video game as stimulus. Cluster analyses depicted that these representations were emotionally linked and socially shared among the players without regard to age or gender. The relation with the designers’ representations about the game was found to be not significant.

Deep supervised, but not unsupervised, models may explain IT cortical representation.

Inferior temporal (IT) cortex in human and nonhuman primates serves visual object recognition. Computational object-vision models, although continually improving, do not yet reach human performance. It is unclear to what extent the internal representations of computational models can explain the IT representation. Here we investigate a wide range of computational model representations (37 in total), testing their categorization performance and their ability to account for the IT representational geometry. The models include well-known neuroscientific object-recognition models (e.g. HMAX, VisNet) along with several models from computer vision (e.g. SIFT, GIST, self-similarity features, and a deep convolutional neural network). We compared the representational dissimilarity matrices (RDMs) of the model representations with the RDMs obtained from human IT (measured with fMRI) and monkey IT (measured with cell recording) for the same set of stimuli (not used in training the models). Better performing models were more similar to IT in that they showed greater clustering of representational patterns by category. In addition, better performing models also more strongly resembled IT in terms of their within-category representational dissimilarities. Representational geometries were significantly correlated between IT and many of the models. However, the categorical clustering observed in IT was largely unexplained by the unsupervised models. The deep convolutional network, which was trained by supervision with over a million category-labeled images, reached the highest categorization performance and also best explained IT, although it did not fully explain the IT data. Combining the features of this model with appropriate weights and adding linear combinations that maximize the margin between animate and inanimate objects and between faces and other objects yielded a representation that fully explained our IT data. Overall, our results suggest that explaining IT requires computational features trained through supervised learning to emphasize the behaviorally important categorical divisions prominently reflected in IT.

Maintenance of youth-like processing protects against false memory in later adulthood

Normal cognitive aging compromises the ability to form and retrieve associations among features of a memory episode. One indicator of this age-related deficit is older adults' difficulty in detecting and correctly rejecting new associations of familiar items. Comparing 28 younger and 30 older adults on a continuous recognition task with word pairs, we found that older adults whose activation patterns deviate less from the average pattern of younger adults while detecting repaired associations show the following: (1) higher overall memory and fewer false recognitions; (2) stronger functional connectivity of prefrontal regions with middle temporal and parahippocampal gyrus; and (3) higher recall and strategic categorical clustering in an independently assessed free recall task. Deviations from the average young-adult network reflected underactivation of frontoparietal regions instead of overactivation of regions not activated by younger adults. We conclude that maintenance of youth-like task-relevant activation patterns is critical for preserving memory functions in later adulthood.