Improved subspace-based and angle-based outlier detections for fuzzy datasets with a real case study

To overcome curse of dimensionality for outlier detecting in high dimensional dataset, axis-parallel subspace (SOD) and angle-based outlier detection (ABOD) methods were presented. These methods are also friendly used distance-based to detect outliers. In this regard, based on the reality of fuzzy data for explaining the world phenomena, this paper introduces an extended version of both methods for fuzzy dataset. First, the basic concepts of both methods are explained. Next we provide two metrics based on Euclidean and analytic distance to measure distance between fuzzy objects; also Cosine similarity measure formula for calculating the cosine of angle between two difference vectors in high-dimensional fuzzy dataset is illustrated. Then the algorithms to determine outliers of fuzzy datasets by using these metrics and Cosine similarity measure, based on ABOD and SOD algorithms, are presented. Some numerical experimental examples are also presented, in which both real and synthesis datasets are used, For a real numerical examination, we have applied proposed algorithms to data from 15 Iranian petrochemical companies in a fully fuzzy environment. The obtained results show the significant properties of the new methods in detecting outliers.

METHOD OF IDENTIFICATION OF OBJECT STATES ACCORDING TO THE RESULTS OF FUZZY MEASUREMENTS OF CONTROLLED PARAMETERS

The subject of consideration is the task of identifying the states of an object based on the results of fuzzy measurements of a set of controlled parameters. The fuzziness of the initial data of the task further complicates it due to the resulting inequality of the controlled parameters. The aim of the study is to develop a method of identifying the states of a fuzzy object using a fuzzy mechanism of logical output taking into account possible differences in the level of information content of its controlled parameters. The method of obtaining the desired result is based on the modification of the known mathematical apparatus for building an expert system of artificial intelligence by solving two subtasks. The first is the development of a method for assessing the informativity of controlled parameters. The second is the development of a method for constructing a mechanism for logical inference of the relative state of an object based on the results of measuring controlled parameters, which provides identification. In the first problem, a method is proposed for estimating the informativity of parameters, free from the known disadvantages of the traditional Kulbak informativity measure. In implementing the method, it is assumed that the range of possible values for each parameter is divided into subbands in accordance with possible states of the object. For each of these states, the function of belonging to the fuzzy values of the corresponding parameter is defined. At the same time, the correct problem of estimating the informativity of a parameter is solved for cases when this parameter is measured accurately or determined fuzzily by its belonging function. The fundamental difference between the proposed logical output mechanism and the traditional one is the refusal to use the production rule base, which ensures the practical independence of the computational procedure from the dimension of the task. To solve the main problem of identifying states, a non-productive approach is proposed, the computational complexity of which practically does not depend on the dimension of the problem (the product of the number of possible states Results.per the number of controlled parameters). The logic output mechanism generates a probability distribution of the system states. In this case, a set of functions of belonging of each parameter to the range of its possible values for each of the states of the object is used, as well as a set of functions of belonging to fuzzy measurement results of each parameter. Conclusions. Thus, a method of identifying the state of fuzzy objects with a fuzzy non-productive output mechanism is proposed, the complexity of which does not depend on the dimension of the task.

Knitting Ourselves Into Being: the Case of Labour and Hip Domesticity on the Social Network Ravelry.com

<p>Thinking of social media participation in terms of doing work may seem a strange proposition. Yet, social network and handicrafts website Ravelry.com requires a great deal of labour from its members. From painstakingly hand-knitting fuzzy objects to photographing, recording and sharing these objects online, Ravelers must supply evidence of their hard work in order to fully participate in this online community. On Ravelry, “writing oneself into being” (Sundén 2002), or performing one’s self through a textual medium, encompasses much more than simply writing. One must knit oneself into being too. Social capital is then accumulated through extensive cataloguing of handmade items. These ‘finished objects’ of knitting and crochet are imbued with affective meaning as tokens of nurturing and gift-giving, consistent with a historicity in which handicrafts like knitting have been associated with gendered care-work. Yet, much of Ravelry’s activity centres on commodity exchange. Displaying commodity ownership is as important as displaying evidence of labour for the accrual of social capital. Recording and displaying domesticity as both acts of labour and acts of consumption fit within a wider trend of hip domesticity, where demonstrating one’s domesticity has become a facet of popular culture. This project examines Ravelry.com’s emphasis on the placement of a physical object between the self and the social network. The thesis argues that this incorporation of material objects into the structure of a social network challenges notions of disembodiment and immateriality. Ravelry.com demonstrates the need for a discussion of social media participant labour which goes beyond the immaterial and affective.</p>

Knitting Ourselves Into Being: the Case of Labour and Hip Domesticity on the Social Network Ravelry.com

<p>Thinking of social media participation in terms of doing work may seem a strange proposition. Yet, social network and handicrafts website Ravelry.com requires a great deal of labour from its members. From painstakingly hand-knitting fuzzy objects to photographing, recording and sharing these objects online, Ravelers must supply evidence of their hard work in order to fully participate in this online community. On Ravelry, “writing oneself into being” (Sundén 2002), or performing one’s self through a textual medium, encompasses much more than simply writing. One must knit oneself into being too. Social capital is then accumulated through extensive cataloguing of handmade items. These ‘finished objects’ of knitting and crochet are imbued with affective meaning as tokens of nurturing and gift-giving, consistent with a historicity in which handicrafts like knitting have been associated with gendered care-work. Yet, much of Ravelry’s activity centres on commodity exchange. Displaying commodity ownership is as important as displaying evidence of labour for the accrual of social capital. Recording and displaying domesticity as both acts of labour and acts of consumption fit within a wider trend of hip domesticity, where demonstrating one’s domesticity has become a facet of popular culture. This project examines Ravelry.com’s emphasis on the placement of a physical object between the self and the social network. The thesis argues that this incorporation of material objects into the structure of a social network challenges notions of disembodiment and immateriality. Ravelry.com demonstrates the need for a discussion of social media participant labour which goes beyond the immaterial and affective.</p>

Testing the systems with fuzzy discrete components

Modern complex systems are build based on heterogeneous components with various interrela tionships, fuzziness, and uncertainty of the laws of functioning of the components and the system. An important class of such systems comprises hybrid intelligent systems, where the components are represented by analytical models of fuzzy objects, artifi cial neural networks, expert systems, etc. The article considers fuzzy discrete devices being, for example, part of hybrid systems. Fuzzy linear automata (FLA) introduced in the article are used as a mathematical model of such components. The problem of test synthesis for FLA used to detect faults in them is discussed. Normal single-stuck faults are permissible faults in FLA. The faults originating from the replacement of some elements of the FLA characteristic matrices with others (from a given set of alternative ones) are also permissible. Test synthesis methods for FLA belonging to the class of m-deterministic and synchronized automata, as well as arbitrary linear automata have been developed. The fi rst two methods are based on reducing the considered problem of solving systems of linear algebraic equations. It should be noted that there is a well-developed mathematical apparatus applying a few eff ective methods for searching for such solutions. The tests synthesized by these methods for m-deterministic and synchronized FLA are sufficiently short and do not exceed the memory depth of the corresponding automata. It is shown that the conditions for an FLA referring to the two fi rst classes mentioned above are not too strict. It is noted that the known methods of test synthesis for linear automata require compliance with much more stringent requirements. The synthesis method for arbitrary FLA also builds short tests

How to choose a fuzzy similarity measure in decision-making?

Purpose The paper is dedicated to the analysis of fuzzy similarity measures in uncertainty analysis in general, and in economic decision-making in particular. The purpose of this paper is to explain how a similarity measure can be chosen to quantify a qualitative description of similarities provided by experts of a given domain, in the case where the objects to compare are described through imprecise or linguistic attribute values represented by fuzzy sets. The case of qualitative dissimilarities is also addressed and the particular case of their representation by distances is presented. Design/methodology/approach The approach is based on measurement theory, following Tversky’s well-known paradigm. Findings A list of axioms which may or may not be satisfied by a qualitative comparative similarity between fuzzy objects is proposed, as extensions of axioms satisfied by similarities between crisp objects. They enable to express necessary and sufficient conditions for a numerical similarity measure to represent a comparative similarity between fuzzy objects. The representation of comparative dissimilarities is also addressed by means of specific functions depending on the distance between attribute values. Originality/value Examples of functions satisfying certain axioms to represent comparative similarities are given. They are based on the choice of operators to compute intersection, union and difference of fuzzy sets. A simple application of this methodology to economy is given, to show how a measure of similarity can be chosen to represent intuitive similarities expressed by an economist by means of a quantitative measure easily calculable. More detailed and formal results are given in Coletti and Bouchon-Meunier (2020) for similarities and Coletti et al. (2020) for dissimilarities.

Smoke recognition network based on dynamic characteristics

Fire is a fierce disaster, and smoke is the early signal of fire. Since such features as chrominance, texture, and shape of smoke are very special, a lot of methods based on these features have been developed. But these static characteristics vary widely, so there are some exceptions leading to low detection accuracy. On the other side, the motion of smoke is much more discriminating than the aforementioned features, so a time-domain neural network is proposed to extract its dynamic characteristics. This smoke recognition network has these advantages:(1) extract the spatiotemporal with the 3D filters which work on dynamic and static characteristics synchronously; (2) high accuracy, 87.31% samples being classified rightly, which is the state of the art even in a chaotic environments, and the fuzzy objects for other methods, such as haze, fog, and climbing cars, are distinguished distinctly; (3) high sensitiveness, smoke being detected averagely at the 23rd frame, which is also the state of the art, which is meaningful to alarm early fire as soon as possible; and (4) it is not been based on any hypothesis, which guarantee the method compatible. Finally, a new metric, the difference between the first frame in which smoke is detected and the first frame in which smoke happens, is proposed to compare the algorithms sensitivity in videos. The experiments confirm that the dynamic characteristics are more discriminating than the aforementioned static characteristics, and smoke recognition network is a good tool to extract compound feature.

Spatial Co-Location Pattern Discovery from Fuzzy Objects

A spatial co-location pattern is a group of spatial objects whose instances are frequently located in the same region. The spatial co-location pattern mining problem has been investigated extensively in the past due to its broad range of applications. In this paper we study this problem for fuzzy objects. Fuzzy objects play an important role in many areas, such as the geographical information system and the biomedical image database. In this paper, we propose two new kinds of co-location pattern mining for fuzzy objects, single co-location pattern mining (SCP) and range co-location pattern mining (RCP), to mining co-location patterns at a membership threshold or within a membership range. For efficient SCP mining, we optimize the basic mining algorithm to accelerate the co-location pattern generation. To improve the performance of RCP mining, effective pruning strategies are developed to significantly reduce the search space. The efficiency of our proposed algorithms as well as the optimization techniques are verified with an extensive set of experiments.