Leaf Clustering Using Circular Densities

By empowering domain experts to perform interactive exploration of large time series datasets, sketch-based query interfaces have revitalized interest in the well-studied problem of time series similarity search. In this new interaction paradigm, recent similarity algorithms (e.g., Qetch, Peax, LineNet) that attempt to capture perceptually relevant features have supplanted older, more straightforward distance measures (e.g., Euclidean, DTW). However, the downside of these algorithms is the resulting difficulty in designing corresponding index structures to support efficient similarity search over large datasets, thus necessitating brute-force search. This demo will showcase Deep Time Series Similarity Search (DTS3), our pluggable indexing pipeline for arbitrary distance measures. DTS3 can automatically train a foundation model for any custom, user-supplied distance measure with no strict constraints (e.g., differentiability), thus enabling fast retrieval via an off-the-shelf vector DBMS. Using our DeepSketch web interface, participants can compare DTS3 to the baseline brute-force versions of several similarity algorithms to see that our approach can achieve much lower latency without sacrificing accuracy when searching over large, real-world time series datasets.

Similarity or distance measures between two time series play an important role in analysis and retrieval of time series database, which is a fundamental problem in time series data mining. Mathematical model is widely used as a representation of time series, but few papers discuss it in similarity measure of time series. In this paper, we propose a Markov model based technique for similarity/distance measures of variable-length time sequences. State space of Markov model is partitioned by hierarchical clustering method, and the information of state-transition is used to represent a time series. The similarity/distance measures of time sequences can be defined as various functions of the difference between their state-transition information, and some widely used distance measures can be considered as our specific cases. In addition, in modeling procedure, the vector sequence in reconstructed phase space is used instead of the original time sequence, which more effectively reflects the dynamical property of time series. Experimental results show that it works well under the strong noise environment, and it is versatile for various applications by its flexible definition.

In this paper, a new method for the problem of shape representation and classification is proposed. In this method, we define a radius function on the contour of the shape which captures for each point of the boundary, attributes of its related internal part of the shape. We call these attributes as "depth" of the point. Depths of boundary points generate a descriptor sequence which represents the shape. Matching of sequences is performed using dynamic programming method and a distance measure is acquired. At last, different classes of shapes are classified using a hierarchical clustering method and the distance measure. The proposed method can analyze features of each part of the shape locally which this leads to the ability of part analysis and insensitivity to local deformations such as articulation, occlusion and missing parts. We show high efficiency of the proposed method by evaluating it for shape matching and classification of standard shape datasets.

The Sphere Exclusion algorithm is a well-known algorithm used to select diverse subsets from chemical-compound libraries or collections. It can be applied with any given distance measure between two structures. It is popular because of the intuitive geometrical interpretation of the method and its good performance on large data sets. This paper describes Directed Sphere Exclusion (DISE), a modification of the Sphere Exclusion algorithm, which retains all positive properties of the Sphere Exclusion algorithm but generates a more even distribution of the selected compounds in the chemical space. In addition, the computational requirement is significantly reduced, thus it can be applied to very large data sets.

Local Government Budgeting for Gender Equality

The success of deep learning in skin lesion classification mainly depends on the ultra-deep neural network and the significantly large training data set. Deep learning training is usually time-consuming, and large datasets with labels are hard to obtain, especially skin lesion images. Although pre-training and data augmentation can alleviate these issues, there are still some problems: (1) the data domain is not consistent, resulting in the slow convergence; and (2) low robustness to confusing skin lesions. To solve these problems, we propose an efficient structural pseudoinverse learning-based hierarchical representation learning method. Preliminary feature extraction, shallow network feature extraction and deep learning feature extraction are carried out respectively before the classification of skin lesion images. Gabor filter and pre-trained deep convolutional neural network are used for preliminary feature extraction. The structural pseudoinverse learning (S-PIL) algorithm is used to extract the shallow features. Then, S-PIL preliminarily identifies the skin lesion images that are difficult to be classified to form a new training set for deep learning feature extraction. Through the hierarchical representation learning, we analyze the features of skin lesion images layer by layer to improve the final classification. Our method not only avoid the slow convergence caused by inconsistency of data domain but also enhances the training of confusing examples. Without using additional data, our approach outperforms existing methods in the ISIC 2017 and ISIC 2018 datasets.

QuestionsWhich clustering algorithms are most effective according to different cluster validity evaluators? Which distance or dissimilarity measure is most suitable for clustering algorithms?LocationHyrcanian forest, Iran (Asia), Virginia region forest, United States (North America), beech forests, Ukraine (Europe).MethodsWe tested 25 clustering algorithms with nine vegetation data sets comprised of three real data sets and six simulated data sets exhibiting different cluster separation values. The clustering algorithms included both hierarchical and non‐hierarchical partitioning. Five evaluators were employed on each cluster solution to evaluate different clustering algorithms. Algorithms were ranked from best to worst on each clustering evaluator for each data set.ResultsThe comparison revealed that the OPTSIL initiated from a Flexible‐β (−0.25) solution achieved particularly good performance. We also found that Ward's method and Flexible‐β (−0.1) implementations were accurate. K‐means with Hellinger distance was superior to Partitioning Around Medoids (PAM) algorithms. Accordance between distance measures and clustering algorithms was also observed. Bray–Curtis dissimilarity combined with a range of clustering algorithms was successful in most cases. Bray–Curtis dissimilarity proved superior to other distance measures for heterogeneous data sets.ConclusionsAll in all, the results demonstrate that choosing the most suitable method before clustering is critical for achieving maximally interpretable clusters. The complexity of vegetation data sets makes this issue even more important. The choice of distance measure had more effect than the choice of clustering method on the quality of results. Our results illustrate that OPTSIL Flexible‐β (−0.1) and OPTPART could prove superior to alternative conventional clustering algorithms when internal evaluation criteria are used to optimize clustering.

Detection of clusters of microcalcifications using a K-nearest neighbour rule with locally optimum distance metrics

Supplementary Material for the manuscript "Drivers of carbon dioxide emissions: An empirical investigation using hierarchical and non- hierarchical clustering methods"

Hierarchical clustering, a traditional clustering method, has been getting attention again. Among several reasons, a credit goes to a recent paper by Dasgupta in 2016 that proposed a cost function that quantitatively evaluates hierarchical clustering trees. An important question is how to combine this recent advance with existing successful clustering methods. In this paper, we propose a hierarchical clustering method to minimize the cost function of clustering tree by incorporating existing clustering techniques. First, we developed an ensemble tree-search method that finds an integrated tree with reduced cost by integrating multiple existing hierarchical clustering methods. Second, to operate on large and arbitrary shape data, we designed an efficient hierarchical clustering framework, called integrating divisive and ensemble-agglomerate (IDEA) by combining it with advanced clustering techniques such as nearest neighbor graph construction, divisive-agglomerate hybridization, and dynamic cut tree. The IDEA clustering method showed better performance in minimizing Dasgupta's cost and improving accuracy (adjusted rand index) over existing cost-minimization-based, and density-based hierarchical clustering methods in experiments using arbitrary shape datasets and complex biology-domain datasets.

Two distances based on permutations are considered to measure the similarity of two time series according to their strength of dependency. The distance measures are used together with different linkages to get hierarchical clustering methods of time series by dependency. We apply these distances to both simulated theoretical and real data series. For simulated time series the distances show good clustering results, both in the case of linear and non-linear dependencies. The effect of the embedding dimension and the linkage method are also analyzed. Finally, several real data series are properly clustered using the proposed method.

Advances in mass spectrometry-based proteomics have yielded a substantial mapping of the tyrosine phosphoproteome and thus provided an important step toward a systematic analysis of intracellular signaling networks in higher eukaryotes. In this study we decomposed an uncharacterized proteomics data set of 481 unique phosphotyrosine (Tyr(P)) peptides by sequence similarity to known ligands of the Src homology 2 (SH2) and the phosphotyrosine binding (PTB) domains. From 20 clusters we extracted 16 known and four new interaction motifs. Using quantitative mass spectrometry we pulled down Tyr(P)-specific binding partners for peptides corresponding to the extracted motifs. We confirmed numerous previously known interaction motifs and found 15 new interactions mediated by phosphosites not previously known to bind SH2 or PTB. Remarkably, a novel hydrophobic N-terminal motif ((L/V/I)(L/V/I)pY) was identified and validated as a binding motif for the SH2 domain-containing inositol phosphatase SHIP2. Our decomposition of the in vivo Tyr(P) proteome furthermore suggests that two-thirds of the Tyr(P) sites mediate interaction, whereas the remaining third govern processes such as enzyme activation and nucleic acid binding.

Color constancy is the ability to measure image features independent of the color of the scene illuminant and is an important topic in color and computer vision. As many color constancy algorithms exist, different distance measures are used to compute their accuracy. In general, these distances measures are based on mathematical principles such as the angular error and Euclidean distance. However, it is unknown to what extent these distance measures correlate to human vision.Therefore, in this paper, a taxonomy of different distance measures for color constancy algorithms is presented. The main goal is to analyze the correlation between the observed quality of the output images and the different distance measures for illuminant estimates. The output images are the resulting color corrected images using the illuminant estimates of the color constancy algorithms, and the quality of these images is determined by human observers. Distance measures are analyzed how they mimic differences in color naturalness of images as obtained by humans.Based on the theoretical and experimental results on spectral and real-world data sets, it can be concluded that the perceptual Euclidean distance (PED) with weight-coefficients (w R = 0.26, w G = 0.70, w B = 0.04) finds its roots in human vision and correlates significantly higher than all other distance measures including the angular error and Euclidean distance.KeywordsDistance MeasureColor SpaceHuman ObserverAngular ErrorHyperspectral DataThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The aim of this study, which was conducted on the basis of the convergence hypothesis, is to reveal the convergence problems of Turkey towards the European Union based on the basic indicators of the youth labor market. For this purpose, a large gender-disaggregated data set has been constructed with 29 observation units consisting of the European Union Average, European Union Countries, and Turkey, using the basic indicators of the youth labor market that point to the future, within Eurostat and Ilostat data. The clustering method, which is one of the advanced statistical techniques, was preferred to determine which countries are similar to each other and which are different from each other within the data set. In this study, where non-hierarchical and hierarchical clustering methods were used together, it was concluded that Turkey diverges from the developed countries of the European Union, such as Denmark, Germany, and the Netherlands, and is similar to countries such as the EU (27), Bulgaria, Czechia, and Italy. Along with this result, this study also reveals remarkable gender differences in the indicators for young men and young women in the youth labor market in Turkey, and that Turkey’s main convergence problem towards the European Union is realized in NEET rates. In this context, this study is completed with suggestions for various policy measures to address convergence problems, such as NEET, unemployment of young women, and low labor force participation rates of young women in Turkey, within the scope of sustainable development goals such as quality education and gender equality.

Community welfare is one of the important points for a region and is also the essence of national development. The welfare of the people in Indonesia is fairly unequal, especially in East Java. To be able to map an area to the welfare of its people in East Java, one way that can be used is to use clustering. The hierarchical clustering method is one of the clustering methods for grouping data. In hierarchical clustering, single linkage, complete linkage, and average linkage methods are suitable methods for grouping data, which will compare the best method to use. The results of the calculation show that the average linkage method with three clusters is the best calculation with a silhouette index value of 0.6054, with the 1st cluster there are 23 regions, namely the city/district with the highest community welfare, the 2nd cluster there are 11 regions, namely cities/districts with moderate social welfare, and in the third cluster there are 4 regions, namely cities/districts with the lowest community welfare.