Probabilistic Similarity Measure Research Articles

Fuzzy Differential Privacy Theory and Its Applications in Subgraph Counting

Transportation networks are essential to the operation of societies and economies. Protecting the privacy of sensitive information is a meaningful conception in sustainable transport when mining the transportation data. In data mining, differential privacy (DP) has provable privacy guarantees for releasing sensitive data by introducing randomness into query results. However, it suffers from significant accuracy loss of outputs when the query has high sensitivity (e.g., triangle counting). The reason is that the range of random perturbation to each query result in DP is too large. It consists of all possible output values for a query that forms a large or even unbounded interval. However, when impose perturbation only in a small neighborhood of the true query result, the similarity measure based on randomness in DP fails. Thereupon, we introduce fuzziness into DP to formulate new models which have smaller disturbance via fuzzy similarity measures. In this article, we establish a novel and general theory of private data analysis, fuzzy differential privacy (FDP). The new theory FDP aims to acquire a more flexible tradeoff between the accuracy of outputs and the privacy-preserving level of data. FDP combines DP with fuzzy set theory by introducing fuzziness into the query results and characterizing similarities between outputs via multiple fuzzy similarity measures. From this perspective, DP can be viewed as a special case of FDP with probabilistic similarity measure. Compared with DP, FDP has three superiorities: 1) most fuzzy similarity measures in FDP support sliding window perturbation strategies we proposed, which refer to perturbation in a small neighborhood of the query results; 2) FDP adds noise to the query results only according to a fraction of all possible neighboring datasets; and 3) the fuzzy similarity with valued in [0,1] quantifies the privacy protection level intuitively. These three points enable more accurate outputs while providing provable and intuitive privacy guarantees. As for subgraph counting, the state-of-the-art method is ladder framework in DP. We illustrate FDP mechanisms by applying them to a common application in subgraph counting–triangle/4-cliques counting. Experiments show that FDP is effective and efficient with smaller output errors than DP.

Three-way multi-attribute decision making under incomplete mixed environments using probabilistic similarity

In recent years, many scholars have explored a variety of methods integrating three-way decision (3WD) and multi-attribute decision making (MADM), which enables the classification and priority ranking of alternatives possible and fully reflects the effectiveness and advantages of 3WD in solving MADM problems. However, few of these methods can effectively deal with the MADM problems with incomplete mixed information that are frequently encountered in real-world situations. This study proposes a three-way MADM method for an incomplete mixed information system (IMIS), where the objective determination of conditional probabilities and utility functions in IMIS without decision label is the pivotal issue. To overcome this issue, we define a probabilistic similarity measure for incomplete mixed information. The probabilistic similarity measure is used to replace the distance measure of classical TOPSIS for estimating the conditional probabilities objectively. The probabilistic similarity class is introduced with arithmetic average method to design a conversion mechanism and obtain the objective relative utility functions of incomplete evaluation values. We then construct a novel 3WD model in IMIS and combine it with two customized ranking principles, to solve the incomplete mixed MADM problems from the perspective of classification and ranking in a more thoughtful and interpretable manner. Our study provides a new perspective for the research on MADM in incomplete mixed information environment. Several examples and experimental comparisons verify the effectiveness and stability of the proposed method. The experiments demonstrate that our method can meet more decision-making requirements and is more accurate and rational in some decision-making scenarios than several prior similar methods.

Effect of Probabilistic Similarity Measure on Metric-Based Few-Shot Classification

In developing a few-shot classification model using deep networks, the limited number of samples in each class causes difficulty in utilizing statistical characteristics of the class distributions. In this paper, we propose a method to treat this difficulty by combining a probabilistic similarity based on intra-class statistics with a metric-based few-shot classification model. Noting that the probabilistic similarity estimated from intra-class statistics and the classifier of conventional few-shot classification models have a common assumption on the class distributions, we propose to apply the probabilistic similarity to obtain loss value for episodic learning of embedding network as well as to classify unseen test data. By defining the probabilistic similarity as the probability density of difference vectors between two samples with the same class label, it is possible to obtain a more reliable estimate of the similarity especially for the case of large number of classes. Through experiments on various benchmark data, we confirm that the probabilistic similarity can improve the classification performance, especially when the number of classes is large.

Global trends assessment of environmental health degradation studies from 1990 to 2018

The impacts of anthropogenic activities (AA) on the natural environment have been one of the most critical and dynamic research areas within the scope of environmental assessment in recent years. This study aimed at investigating the probabilistic affinity index between environmental health degradation (EHD) and AA globally. In this research, a global scientometric assessment of 607 academic research articles between 1991 and 2018 on EHD and other factors such as human and natural activities is presented in this study to characterise the intellectual dynamics by visualising and identifying the strength, evolution of the co-citation network, collaboration network and trends. An in-depth analysis of EHD and AA research was conducted to better understand the global trends and directions that have emerged over the past 28 years in this field. The results reveal that the number of publication in the field of EHD has increased significantly and it has become noticeably an interdisciplinary research in recent years. It further revealed that most of the authors and their affiliated institutions in this research area are from USA, China, Brazil, UK, Australia, Canada, France, India, Italy and Germany. Findings from the study show that the published research in the field has increased in number, tending mostly towards an interdisciplinary research subject. The most pressing link between EHD and human activities, as represented in existing literature, includes the consequences of EHD for biodiversity and ecosystems sustainability, and emission reductions in the environmental sector. This study reveals that the probabilistic similarity measures of EHD and AA are strongly correlated with each other. The study concluded that there is a global increase in environmental health degradation research; noticeable research outcomes are mostly witnessed in developed countries compared to the developing nations.

A Uniqueness-Driven Similarity Measure for Automated Competitor Identification

Uniqueness is an important source of competitive advantage and a salient aspect for firms identifying competitors and market structure. While marketing research often includes uniqueness as an important aspect of product positioning and product strategy, the existing literature has offered little guidance on operationalising this notion for use in the competitor identification process. This paper proposes a probabilistic similarity measure to quantify a competitive landscape where uniqueness is a key driver of competition. The proposed measure, when used with readily available data and combined with existing clustering algorithms, enables automation of the competitor identification process. Empirical experiments are used to validate the proposed measure. These experiments show that marketers can use readily available data, including social media tags and geographical proximity data, to reveal the same insight as is gathered when using the more laborious and time-consuming approach of traditional consumer surveys.

Novel Adaptive Clustering Algorithms Based on a Probabilistic Similarity Measure Over Atanassov Intuitionistic Fuzzy Set

This paper presents a novel probabilistic similarity measure (PSM) for Atanassov intuitionistic fuzzy sets. It then exploits PSM to propose an adaptive probabilistic similarity degree and develops the novel probabilistic $\lambda$ -cutting algorithm for clustering. Further, the probabilistic distance measure (obtained from the PSM) is used to develop a new clustering technique, which we have named “probabilistic intuitionistic fuzzy c-mean (PIFCM) algorithm”. Simulation experiments have been conducted over a variety of datasets including UCI machine learning datasets and real-world car dataset. The results obtained have been thoroughly compared with other well-known clustering techniques such as fuzzy c-mean (FCM), intuitionistic fuzzy c-mean, association coefficient method, and $\lambda$ -cutting method. Based upon the experimental results, it can be concluded that our probabilistic $\lambda$ -cutting algorithm and PIFCM algorithm outperform their existing counterparts.

Generic multivariate model for color texture classification in RGB color space

This paper presents a new method for modeling magnitudes of dual-tree complex wavelet coefficients, in the context of color texture classification. Based on the characterization of dependency between RGB color components, Gaussian copula associated with Generalized Gamma marginal function is proposed to design the multivariate generalized Gamma density (MG\(\Gamma \)D) modeling. MG\(\Gamma \)D has the advantages of genericity in terms of fitting over a variety of existing joint models. On the one hand, the generalized Gamma density function offers free-shape parameters to characterize a wide range of heavy-tailed densities, i.e., the genericity. On the other hand, the inter-component, inter-band dependency is captured by the Gaussian Copula which offers adapted flexibility. Moreover, this model leads to a closed form for the probabilistic similarity measure in terms of parameters, i.e., Kullback–Leibler divergence. By exploiting the separability between the copula and the marginal spaces, the closed form enables us to minimize the computational time needed to measure the discrepancy between two Multivariate Generalized Gamma densities in comparison to other models which imply using a Monte Carlo method characterized by an expensive time computing. For evaluating the performance of our proposal, a K-nearest neighbor (KNN) classifier is then used to test the classification accuracy. Experiments on different benchmarks using color texture databases are conducted to highlight the effectiveness of the proposed model associated to the Kullback–Leibler divergence.

3D soft-tissue tracking using spatial-color joint probability distribution and thin-plate spline model

Visual tracking techniques based on stereo endoscope are developed to measure tissue motion in robot-assisted minimally invasive surgery. However, accurate 3D tracking of tissue surfaces remains challenging due to complicated deformation, poor imaging conditions, specular reflections and other dynamic effects during surgery. This study employs a robust and efficient 3D tracking scheme with two independent recursive processes, namely kernel-based inter-frame motion estimation and model-based intra-frame 3D matching. In the first process, target region is represented in joint spatial-color space for robust estimation. By defining a probabilistic similarity measure, a mean-shift-based iterative algorithm is derived for location of the target region in a new image. In the second process, the thin-plate spline model is used to fit the 3D shape of tissue surfaces around the target region. An iterative algorithm based on an efficient second-order minimization technique is derived to compute optimal model parameters. The two processes can be computed in parallel. Their outputs are combined to recover 3D information about the target region. The performance of the proposed method is validated using phantom heart videos and in vivo videos acquired by the daVinci® surgical robotic platform and a synthesized data set with known ground truth.

Characterization and analysis of sales data for the semiconductor market: An expert system approach

Chip purchasing policies of the Original Equipment Manufacturers (OEMs) of laptop computers are characterized by similarity measures and probabilistic rules. Our main goal is to build an expert system for predicting purchasing behavior in the semiconductor market. The probabilistic rules and similarity measures are extracted from data of products bought by the OEMs in the semiconductor market over twenty quarters. We present the data collected and different qualitative data mining approaches to analyze and extract rules from the data that best characterize the purchasing behavior of the OEMs. Our analysis of the similar product selection shows that there are two main groups of OEMs buying similar products. Using our probabilistic rules, we obtain an average score of approximately 95% reconstructing quarterly data for a one year window.

SMETANA: Accurate and Scalable Algorithm for Probabilistic Alignment of Large-Scale Biological Networks

In this paper we introduce an efficient algorithm for alignment of multiple large-scale biological networks. In this scheme, we first compute a probabilistic similarity measure between nodes that belong to different networks using a semi-Markov random walk model. The estimated probabilities are further enhanced by incorporating the local and the cross-species network similarity information through the use of two different types of probabilistic consistency transformations. The transformed alignment probabilities are used to predict the alignment of multiple networks based on a greedy approach. We demonstrate that the proposed algorithm, called SMETANA, outperforms many state-of-the-art network alignment techniques, in terms of computational efficiency, alignment accuracy, and scalability. Our experiments show that SMETANA can easily align tens of genome-scale networks with thousands of nodes on a personal computer without any difficulty. The source code of SMETANA is available upon request. The source code of SMETANA can be downloaded from http://www.ece.tamu.edu/~bjyoon/SMETANA/.

Fast alignment and comparison of RNA structures

To recognize remote relationships between RNA molecules, one must be able to align structures without regard to sequence similarity. We have implemented a method, which is swift [O(n(2))], sensitive and tolerant of large gaps and insertions. Molecules are broken into overlapping fragments, which are characterized by their memberships in a probabilistic classification based on local geometry and H-bonding descriptors. This leads to a probabilistic similarity measure that is used in a conventional dynamic programming method. Examples are given of database searching, the detection of structural similarities, which would not be found using sequence based methods, and comparisons with a previously published approach. Source code (C and perl) and binaries for linux are freely available at www.zbh.uni-hamburg.de/fries.

A novel approach for effective web page classification

With the exponential increase in volume of the WWW every day, web page classification has become tedious. Since with no quality data there is no quality mining results, it is worth to emphasise on fine tuning the data for classification, rather than improving the classifiers themselves. This paper investigates the methods for improving web page classification by feature extraction, selection and data tuning. This paper also proposes a new classification model for web page classification called a probabilistic web page classifier (PWPC). It is based on a probabilistic framework and attribute-value similarity measure (AVS). The proposed method is tested on a benchmarking dataset, WebKB and the performance of PWPC on the fine tuned web pages has exhibited significant accuracy over the traditional machine learning classifiers.

A Nonlocal SAR Image Denoising Algorithm Based on LLMMSE Wavelet Shrinkage

We propose a novel despeckling algorithm for synthetic aperture radar (SAR) images based on the concepts of nonlocal filtering and wavelet-domain shrinkage. It follows the structure of the block-matching 3-D algorithm, recently proposed for additive white Gaussian noise denoising, but modifies its major processing steps in order to take into account the peculiarities of SAR images. A probabilistic similarity measure is used for the block-matching step, while the wavelet shrinkage is developed using an additive signal-dependent noise model and looking for the optimum local linear minimum-mean-square-error estimator in the wavelet domain. The proposed technique compares favorably w.r.t. several state-of-the-art reference techniques, with better results both in terms of signal-to-noise ratio (on simulated speckled images) and of perceived image quality.

De-duplication of aggregation authority files

This paper presents PACE Programmable Authority Control Engine, an authority control tool conceived to maintain 'aggregation authority files'. These are obtained as continuous aggregations of records originating from a variable set of information systems with heterogeneous and duplicated content. To facilitate record deduplication in the presence of such heterogeneity and dynamicity, PACE user interfaces enable an iterative curation process, where data curators can: i configure algorithms for the identification of record duplicates; ii open work sessions where algorithm configurations can be run and evaluated; iii merge the identified record duplicates to disambiguate the authority file and iv repeat this cycle several times. PACE supports a tunable probabilistic similarity measure and performs record matching with a customisable variation of the sorted neighbourhood heuristic. Finally, it addresses the underlying performance and scalability issues by exploiting multi-core parallel processing and Cassandra's storage systems, to support I/O performances that scale up linearly with the number of records.

P-value based visualization of codon usage data

Two important and not yet solved problems in bacterial genome research are the identification of horizontally transferred genes and the prediction of gene expression levels. Both problems can be addressed by multivariate analysis of codon usage data. In particular dimensionality reduction methods for visualization of multivariate data have shown to be effective tools for codon usage analysis. We here propose a multidimensional scaling approach using a novel similarity measure for codon usage tables. Our probabilistic similarity measure is based on P-values derived from the well-known chi-square test for comparison of two distributions. Experimental results on four microbial genomes indicate that the new method is well-suited for the analysis of horizontal gene transfer and translational selection. As compared with the widely-used correspondence analysis, our method did not suffer from outlier sensitivity and showed a better clustering of putative alien genes in most cases.

New strategy for the representation and the integration of biomolecular knowledge at a cellular scale.

The combination of sequencing and post-sequencing experimental approaches produces huge collections of data that are highly heterogeneous both in structure and in semantics. We propose a new strategy for the integration of such data. This strategy uses structured sets of sequences as a unified representation of biological information and defines a probabilistic measure of similarity between the sets. Sets can be composed of sequences that are known to have a biological relationship (e.g. proteins involved in a complex or a pathway) or that share similar values for a particular attribute (e.g. expression profile). We have developed a software, BlastSets, which implements this strategy. It exploits a database where the sets derived from diverse biological information can be deposited using a standard XML format. For a given query set, BlastSets returns target sets found in the database whose similarity to the query is statistically significant. The tool allowed us to automatically identify verified relationships between correlated expression profiles and biological pathways using publicly available data for Saccharomyces cerevisiae. It was also used to retrieve the members of a complex (ribosome) based on the mining of expression profiles. These first results validate the relevance of the strategy and demonstrate the promising potential of BlastSets.

Relational object recognition from large structural libraries

This paper presents a probabilistic similarity measure for object recognition from large libraries of line-patterns. We commence from a structural pattern representation which uses a nearest neighbour graph to establish the adjacency of line-segments. Associated with each pair of line-segments connected in this way is a vector of Euclidean invariant relative angle and distance ratio attributes. The relational similarity measure uses robust error kernels to compare sets of pairwise attributes on the edges of a nearest neighbour graph. We use the relational similarity measure in a series of recognition experiments which involve a library of over 2500 line-patterns. A sensitivity study reveals that the method is capable of delivering a recognition accuracy of 94%. A comparative study reveals that the method is most effective when either a Gaussian kernel or Huber's robust kernel is used to weight the attribute relations. Moreover, the method consistently outperforms the standard and the quantile Hausdorff distance.

A Continuous Probabilistic Framework for Image Matching

In this paper we describe a probabilistic image matching scheme in which the image representation is continuous and the similarity measure and distance computation are also defined in the continuous domain. Each image is first represented as a Gaussian mixture distribution and images are compared and matched via a probabilistic measure of similarity between distributions. A common probabilistic and continuous framework is applied to the representation as well as the matching process, ensuring an overall system that is theoretically appealing. Matching results are investigated and the application to an image retrieval system is demonstrated.

A Bayesian similarity measure for deformable image matching

Abstract We propose a probabilistic similarity measure for direct image matching based on a Bayesian analysis of image deformations. We model two classes of variation in object appearance: intra-object and extra-object. The probability density functions for each class are then estimated from training data and used to compute a similarity measure based on the a posteriori probabilities. Furthermore, we use a novel representation for characterizing image differences using a deformable technique for obtaining pixel-wise correspondences. This representation, which is based on a deformable 3D mesh in XYI-space, is then experimentally compared with two simpler representations: intensity differences and optical flow. The performance advantage of our deformable matching technique is demonstrated using a typically hard test set drawn from the US Army's FERET face database.

Bayesian face recognition

We propose a new technique for direct visual matching of images for the purposes of face recognition and image retrieval, using a probabilistic measure of similarity, based primarily on a Bayesian (MAP) analysis of image differences. The performance advantage of this probabilistic matching technique over standard Euclidean nearest-neighbor eigenface matching was demonstrated using results from DARPA's 1996 “FERET” face recognition competition, in which this Bayesian matching alogrithm was found to be the top performer. In addition, we derive a simple method of replacing costly computation of nonlinear (on-line) Bayesian similarity measures by inexpensive linear (off-line) subspace projections and simple Euclidean norms, thus resulting in a significant computational speed-up for implementation with very large databases.