Narrow Results

In this paper we present a voting scheme for fuzzy cluster algorithms. This voting method allows us to combine several runs of cluster algorithms resulting in a common partition. This helps us to tackle the problem of choosing the appropriate clustering method for a data set where we have no a priori information about it. We mathematically derive the algorithm from theoretical considerations. Experiments show that the voting algorithm finds structurally stable results. Several cluster validity indexes show the improvement of the voting result in comparison to simple fuzzy voting.

In this paper, we show that simple edge characteristics in images, when judiciously combined, can result in improved scene and object classification. Unlike existing methods that require a large number of training samples and complex learning schemes, our method discovers simple edge properties. We introduce three sets of edge properties, namely, centroid, compactness and aspect ratio of edges in the image. The combinations of these edge properties are used to discriminate among images in each class. A class representative is calculated for each class according to the average percentage of edges that satisfy the property of a particular class. This percentage for an unknown image is compared to the class representative to assign a label to it. It is shown that this simple edge properties-based method outperforms some of the state-of-the-art results on scene and object classification on standard databases.

Shape extraction aims to detect and localize objects via the shape information. The paper presents a novel voting scheme that can extract partially occluded objects under cluttered environments using a single shape. It works by jointly figuring out the boundaries and resolving the geometric configurations. To model the missing part lead by occlusion, we discretize the shape template into a set of its subpart, named portions. Our representation of shape template is through a set of portion together with their interconnections. Instead of forming a fully connected network, our interconnections make the portions consistent with the chain along the boundary of shape template. Based on the representation, we formulate an auto-locked objective function that contains both the unary and pairwise terms and balances the effects of missing parts. Min-sum voting scheme with strategy driven by bottom–up information is then proposed to minimize the objective function. Conducted experiments show that proposed algorithm is promising for shape extraction with occlusion and noisy backgrounds and allows the non-rigid deformations.

This paper describes the investigation results about the usage of shallow (limited by few layers only) convolutional neural networks (CNNs) to solve the video-based gender classification problem. Different architectures of shallow CNN are proposed, trained and tested using balanced and unbalanced static image datasets. The influence of diverse voting over confidences methods, applied for frame-by-frame gender classification of the video stream, is investigated for possible enhancement of the classification accuracy. The possibility of the grouping of shallow networks into ensembles is investigated; it has been shown that the accuracy may be more improved with the further voting of separate shallow CNN classification results inside an ensemble over a single frame or different ones.

Due to the increasing severity of the COVID-19 pandemic, timely screening and diagnosis of infections are essential. Since cough is a common symptom of COVID-19, an AI-assisted cough classification scheme is designed in this paper to diagnose COVID-19 infection. To reduce the labeling efforts by human experts, a semi-supervised learning with voting scheme using a triple-classifier model is proposed for the COVID-19 cough classification. This work aims to improve the accuracy of the classification. Initially, the data pre-processing scheme is executed by performing data cleaning, resampling, and data enhancement so as to improve the audio quality before training. The pre-training scheme is then performed by using a few numbers of COVID-19 cough data with labeling. Then we modify a well-known self-supervised learning model, SimCLR, to a semi-supervised learning-based SimCLR-like model, which uses three different loss functions to fine-tune three training models for cough classification. Finally, a voting scheme is performed based on the classification results of the three cough classifiers so as to enhance the accuracy of the cough classification for COVID-19. The experiment results illustrate that the proposed scheme can achieve 85% accuracy, which outperforms the existing semi-supervised learning-based classification schemes.

When optical character recognition (OCR) devices process the same page image, they generate similar text strings. Differences are due to recognition errors. A page of text rarely contains long repeated substrings; therefore, N strings generated by OCR devices can be quickly matched by detecting long common substrings. An algorithm for matching an arbitrary number of strings based on this principle is presented. Although its worst-case performance is O(Nn²), its performance in practice has been observed to be O(Nn log n), where n is the length of a string. This algorithm has been successfully used to study OCR errors, to determine the accuracy of OCR devices, and to implement a voting algorithm.

This paper presents a multi-stage software design approach for fault-tolerance. In the first stage, a formalism is introduced to represent the behavior of the system by means of a set of assertions. This formalism enables an execution tree (ET) to be generated where each path from the root to the leaf is, in fact, a well-defined formula. During the automatic generation of the execution tree, properties like completeness and consistency of the set of assertions can be verified and consequently design faults can be revealed. In the second stage, the testing strategy is based on a set of WDFs. This set represents the structural deterministic test for the model of the software system and provides a framework for the generation of a functional deterministic test for the code implementation of the model. This testing strategy can reveal the implementation faults in the program code. In the third stage, the fault-tolerance of the software system against hardware failures is improved in a way such that the design and implementation features obtained from the first two stages are preserved. The proposed approach provides a high level of user-transparency by employing object-oriented principles of data encapsulation and polymorphism. The reliability of the software system against hardware failures is also evaluated. A tool, named Software Fault-Injection Tool (SFIT), is developed to estimate the reliability of a software system.

This paper examines 781 proposals sponsored by minority shareholders of Chinese listed companies since the Company Law was revised in 2005, which made proposals more accessible to smaller shareholders. I find that the shareholder structure is essential for minority shareholders to target a company. When the controlling ownership is smaller and minority ownership is relatively larger, a company is more likely to be targeted by minority shareholders’ proposals (MSPs). There is also evidence that minority shareholders target poor performers, but that only happens when the shareholder structure is optimal for minority shareholders to initiate activism. Power, in the form of ownership and identity, dominates the probability of activism behavior. The passage of MSPs heavily depends on whether the proposal is friendly to the management, the identity of the sponsor, and the issues involved. Being targeted by MSPs are often followed by forced top management turnover, but there are no performance improvements in the medium term. Minority shareholders’ activism is more of a game of power.

This paper aims to give a global vision concerning the state of the art of studies on 13 power indices and to establish which of them are more suitable for describing the real situations which are, from time to time, taken into consideration. In such contexts, different comparisons have been developed in terms of properties, axiomatic grounds and so on. This analysis points out various open problems.

This paper presents a review of literature on simple games and highlights various open problems concerning such games; in particular, weighted games and power indices.

The paper presents some issues currently under studying in the field of Cooperative Games. The related open problems are also mentioned.

This paper proposes a brief review of the use of power indices in the corporate governance literature. Without losing sight of the field of application, it places the emphasis on the game-theoretic aspects of this research and on the issues that arise in this framework. It should be noted that the views presented in this paper are not necessarily those of the IMF.

The aim of this work is to give a characterization of the Shapley–Shubik and the Banzhaf power indices for (3,2)-simple games. We generalize to the set of (3,2)-simple games the classical axioms for power indices on simple games: transfer, anonymity, null player property and efficiency. However, these four axioms are not enough to uniquely characterize the Shapley–Shubik index for (3,2)-simple games. Thus, we introduce a new axiom to prove the uniqueness of the extension of the Shapley–Shubik power index in this context. Moreover, we provide an analogous characterization for the Banzhaf index for (3,2)-simple games, generalizing the four axioms for simple games and adding another property.

All voting procedures are susceptible to give rise, if not to paradoxes, at least to violations of some democratic principles. In this paper, we evaluate and compare the propensity of various voting rules -belonging to the class of scoring rules- to satisfy two versions of the majority principle. We consider the asymptotic case where the numbers of voters tends to infinity and, for each rule, we study with the help of Monte Carlo methods how this propensity varies as a function of the number of candidates.

How can networking affect the turnout in an election? We present a simple model to explain turnout as a result of a dynamic process of formation of the intention to vote within Erdös–Rényi networks. Citizens have fixed preferences for one of two parties and are embedded in a social network. They decide whether or not to vote on the basis of the attitude of their immediate contacts. They may simply follow the behavior of the majority (followers) or make an adaptive local calculus of voting (calculators). So they have the intention of voting either when the majority of their neighbors are willing to vote too, or when they perceive in their social neighborhood that elections are "close". We study the long-run average intention to vote, interpreted as the actual turnout observed in an election. Depending on the values of the average connectivity and the probability of behaving as a follower/calculator, the system exhibits monostability (zero turnout), bistability (zero and moderate/high turnout) or tristability (zero, moderate and high turnout). By obtaining realistic turnout rates for a wide range of values of both parameters, our model suggests a mechanism behind the observed relevance of social networks in recent elections.

Social choice theory is the study of decision theory on how to aggregate separate preferences into group's rational preference. It has wide applications, especially on the design of voting rules, and brings far-reaching influence on the development of modern political science and welfare economics. With the advent of the information age, social choice theory finds its up-to-date application on designing effective Metasearch engines. Metasearch engines provide effective searching by combining the results of multiple source search engines that make use of diverse models and techniques. In this work, we analyze social choice algorithms in a graph-theoretic approach. In addition to classical social choice algorithms, such as Borda and Condorcet, we study one special type of social choice algorithms, elimination voting, to tackle Metasearch problem. Some new algorithms are proposed and examined in the fusion experiment on TREC data. It shows that these elimination voting algorithms achieve satisfied performance when compared with Borda algorithm.

In this paper, we develop a method based on the idea of pairwise voting to rank projects or candidates and incorporate in the ranking process how strongly the referees/voters feel about the comparisons they make. Voting is a modified form of ranking and all the votes are equally important. However, there are situations similar to voting in which the votes are not just ordinal but each voter expresses an intensity of preference for the different candidates, e.g., ranking projects for funding. We show that our method yields the same results as ordinal voting in large populations when the intensity of preferences becomes extreme. Voting with intensity of preferences does not violate democracy but soften the stand of voters and allows for consideration of the diversity of issues involved in voting.

In this paper, the effectiveness of three different operating strategies applied to the Fuzzy ARTMAP (FAM) neural network in pattern classification tasks is analyzed and compared. Three types of FAM, namely average FAM, voting FAM, and ordered FAM, are formed for experimentation. In average FAM, a pool of the FAM networks is trained using random sequences of input patterns, and the performance metrics from multiple networks are averaged. In voting FAM, predictions from a number of FAM networks are combined using the majority-voting scheme to reach a final output. In ordered FAM, a pre-processing procedure known as the ordering algorithm is employed to identify a fixed sequence of input patterns for training the FAM network. Three medical data sets are employed to evaluate the performances of these three types of FAM. The results are analyzed and compared with those from other learning systems. Bootstrapping has also been used to analyze and quantify the results statistically.

This paper reports on the development of a Named Entity Recognition (NER) system in Bengali by combining the outputs of the three classifiers, namely Maximum Entropy (ME), Conditional Random Field (CRF) and Support Vector Machine (SVM). A part of the Bengali news corpus developed from the web-archive of a leading Bengali newspaper has been manually annotated with the four major named entity (NE) tags, namely Person name, Location name, Organization name and Miscellaneous name. We have also used the annotated corpus of the IJCNLP-08 NER Shared Task for South and South East Asian Languages (NERSSEAL). An appropriate tag conversion routine has been developed in order to convert the fine-grained NE tagged NERSSEAL corpus to the form, tagged with the coarse-grained NE tagset of four tags. The system makes use of the different contextual information of the words along with the variety of orthographic word-level features that are helpful in predicting the four NE classes. In this work, we have considered language independent features as well as the language dependent features extracted from the various language specific resources. Lexical context patterns, which are generated from an unlabeled corpus of 10 million wordforms using an active learning technique, have been used for developing a baseline NER system as well as the features of the classifiers in order to improve their performance. A number of post-processing techniques have been used in order to improve the performance of the classifiers. Finally, the classifiers are combined together into a multiengine NER system using three weighted voting techniques. The system has been trained and tested with the datasets of 272K wordforms and 35K wordforms, respectively. Experimental results show the effectiveness of the proposed approach with the overall average Recall, Precision and F-Score values of 93.81%, 92.18% and 92.98%, respectively. The proposed system also outperforms the three other existing Bengali NER systems. The language independent versions of the ME, CRF and SVM based NER systems have been evaluated for the four other popular Indian languages, namely Hindi, Telugu, Oriya and Urdu, with the datasets obtained from the NERSSEAL shared task data. The SVM based system yielded the best performance with the F-Score values of 76.35%, 72.65%, 69.34% and 65.66% for Hindi, Telugu, Oriya and Urdu, respectively.

The aim of this work is to give a characterization of the Shapley–Shubik and the Banzhaf power indices for (3,2)-simple games. We generalize to the set of (3,2)-simple games the classical axioms for power indices on simple games: transfer, anonymity, null player property and efficiency. However, these four axioms are not enough to uniquely characterize the Shapley–Shubik index for (3,2)-simple games. Thus, we introduce a new axiom to prove the uniqueness of the extension of the Shapley–Shubik power index in this context. Moreover, we provide an analogous characterization for the Banzhaf index for (3,2)-simple games, generalizing the four axioms for simple games and adding another property.