Search results

Twin support vector machine (TSVM) is an effective machine learning technique. However, the TSVM model does not consider the influence of different data samples on the optimal hyperplane, which results in its sensitivity to noise. To solve this problem, this study proposes a twin support vector machine model based on fuzzy systems (FSTSVM).

This study designs an effective fuzzy membership assignment strategy based on fuzzy systems. It describes the relationship between the three inputs and the fuzzy membership of the sample by defining fuzzy inference rules and then exports the fuzzy membership of the sample. Combining this strategy with TSVM, the FSTSVM is proposed. Moreover, to speed up the model training, this study employs a coordinate descent strategy with shrinking by active set. To evaluate the performance of FSTSVM, this study conducts experiments designed on artificial data sets and UCI data sets.

The experimental results affirm the effectiveness of FSTSVM in addressing binary classification problems with noise, demonstrating its superior robustness and generalization performance compared to existing learning models. This can be attributed to the proposed fuzzy membership assignment strategy based on fuzzy systems, which effectively mitigates the adverse effects of noise.

This study designs a fuzzy membership assignment strategy based on fuzzy systems that effectively reduces the negative impact caused by noise and then proposes the noise-robust FSTSVM model. Moreover, the model employs a coordinate descent strategy with shrinking by active set to accelerate the training speed of the model.

The purpose of this study is to address the challenge of task allocation in multi-robot systems by getting the minimum overall task completion time and task allocation scheme while also minimizing robot energy consumption. This study aims to move away from traditional centralized methods and validate a more scalable distributed approach.

This paper proposes a distributed algorithm for the multi-robot task allocation problem, aimed at getting the minimum task completion time along with the task allocation scheme. The algorithm operates based on local interaction information rather than global information. By using the Consensus-Based Auction Algorithm (CBAA), it seeks to effectively minimize energy consumption without affecting the minimum completion time required for overall task allocation.

The proposed distributed algorithm successfully reduces robot energy consumption while effectively obtaining the shortest overall task completion time and corresponding task allocation scheme. Numerical simulations conducted using MATLAB software demonstrated its superior performance, and empirical testing on the Turtlebot3-Burger robot platform further substantiated these findings.

The original contribution of this study lies in the development of an enhanced distributed task allocation strategy using CBAA to improve efficiency in multi-robot environments. Its value extends to applications that require rapid and resource-aware coordination, such as automated logistics or search-and-rescue operations.

This study aims to take a public goods approach to understand relationships between collecting and contributing knowledge to an online knowledge sharing portal (KSP), mental model processing and outcomes at the individual and collective levels.

This study reports on a survey (N = 602) among tax professionals, examining the perceived individual and collective benefits and costs associated with collecting and contributing knowledge. Hypotheses were tested using structural equation modeling.

Collecting and contributing knowledge led to considerable mental model processing of the knowledge. That in turn significantly influenced (primarily) individual and (some) collective costs and benefits. Results varied by the kinds of knowledge sharing. Whether directly from knowledge sharing, or mediated through mental modeling, the perceived costs and benefits may be internalized as an individual good rather than being interpreted at the collective level as a public good.

The study is situated in the early stages of a wiki-type online KSP. A focus on the learning potential of the system could serve to draw in new users and contributors, heightening perceptions of the public goods dimension of a KSP.

A focus on the learning potential of the system could serve to draw in new users, and thus the number of subsequent contributors, heightening perceptions of the collective, public goods dimension of a KSP.

This study explores how knowledge sharing and mental model processing are directly and indirectly associated with individual and collective costs and benefits. As online knowledge sharing is both an individual and public good, costs and benefits must be considered from both perspectives.

This paper aims to propose a diverless weld bead maintenance welding technology to prevent the leakage of subsea oil and gas pipeline and solve the key problems in the maintenance of subsea pipeline.

Based on the analysis of the cross-section of the fillet weld, the multi-layer and multi-pass welding path planning of the submarine pipeline sleeve fillet weld is studied, and thus a multi-layer and multi-pass welding path planning strategy is proposed. A welding seam filling method is designed, and the end position of the welding gun is planned, which provides a theoretical basis for the motion control of the maintenance system.

The trajectory planning and adjustment of multi-layer and multi-pass fillet welding and the motion stability control of the rotating mechanism are realized.

It provides the basis for the prototype design of the submarine pipeline maintenance and welding robot system, and also lays the foundation for the in-depth research on the intelligent maintenance system of submarine pipeline.

The maintenance of diverless subsea pipeline is a new type of maintenance method, which can solve the problem of large amount of subsea maintenance work with high efficiency.

With the increasing spread of ontologies in various domains, units have gradually become an essential part of ontologies and units ontologies have been developed to offer a better expression ability for the practical usage. From the perspectives of architecture, comparison and reuse, the purpose of this paper is to provide a comprehensive survey on four mainstream units ontologies: quantity-unit-dimension-type, quantities, units, dimensions and values, ontology of units of measure and units ontology (UO) of the open biomedical ontologies, in order to address well the state of the art and the reuse strategies of the UO.

An architecture of units ontologies is presented, in which the relations between key factors (i.e. units of measure, quantity and dimension) are discussed. The criteria for comparing units ontologies are developed from the perspectives of organizational structure, pattern design and application scenario. Then, the authors compare four typical units ontologies based on the proposed comparison criteria. Furthermore, how to reuse these units ontologies is discussed in materials science domain by utilizing two reuse strategies of partial reference and complete reference.

Units ontologies have attracted high attention in the scientific domain. Based on the comparison of four popular units ontologies, this paper finds that different units ontologies have different design features from the perspectives of basis structure, units conversion and axioms design; a UO is better to be applied to the application areas that satisfy its design features; and many challenges remain to be done in the future research of the UO.

This paper makes an extensive review on units ontologies, by defining the comparison criteria and discussing the reuse strategies in the materials domain. Based on this investigation, guidelines are summarized for the selection and reuse of units ontologies.

In the current economic scenario characterized by turbulence, innovation is a requisite for company's growth. The innovation activities are implemented through the realization of innovative project. This paper aims to prospect the promising opportunities coming from the application of Machine Learning (ML) algorithms to project risk management for organizational innovation, where a large amount of data supports the decision-making process within the companies and the organizations.

Moving from a structured literature review (SLR), a final sample of 42 papers has been analyzed through a descriptive, content and bibliographic analysis. Moreover, metrics for measuring the impact of the citation index approach and the CPY (Citations per year) have been defined. The descriptive and cluster analysis has been realized with VOSviewer, a tool for constructing and visualizing bibliometric networks and clusters.

Prospective future developments and forthcoming challenges of ML applications for managing risks in projects have been identified in the following research context: software development projects; construction industry projects; climate and environmental issues and Health and Safety projects. Insights about the impact of ML for improving organizational innovation through the project risks management are defined.

The study have some limitations regarding the choice of keywords and as well the database chosen for selecting the final sample. Another limitation regards the number of the analyzed papers.

The analysis demonstrated how much the use of ML techniques for project risk management is still new and has many unexplored areas, given the increasing trend in annual scientific publications. This evidence represents an opportunities for supporting the organizational innovation in companies engaged into complex projects whose risk management become strategic.

The purpose of this paper is to compare public preferences for investment and spending on non-automobile infrastructures (mass transit and bicycling) to preferences for new roads and the repair of current highways. The study explores the factors that explain preferences for non-automobile infrastructure using a three-factor model including self-interest (personal transportation benefits), concern for community-wide benefits (political beliefs), and concern for the economic impact. The study uses a case study of the urban context of the Hampton Roads region of Southeastern Virginia (USA).

The analysis uses data from a 2013 telephone survey of urban residents in the Hampton Roads area. Survey respondents were asked to identify their two investment priorities from four options: repairing existing roads, bridges, and tunnels; constructing new or expanding roads, bridges, and tunnels; expanding mass transit; and expanding bicycle routes and improving bike safety.

Repairing existing highway infrastructure is the most popular spending priority (66 percent of residents). There is as much support (46 percent) for investing in non-automobile infrastructure as for investing in new roads, bridges, and tunnels. Significant predictors of support for non-automobile infrastructure, using the three-factor model, are: length of commute time, self-identification as liberal, use of light rail, and a belief that light rail contributes to economic development.

The study examines public preferences for both non-traditional and traditional transportation infrastructure investments. It highlights the factors that contribute to public support for different transportation spending options.

Starting from the premises that Internet of Things (IoT) applications can be used in museums as an aid to visiting systems, the purpose of this paper is to see how recommendation systems can be developed to provide advanced services to museum visitors.

The research methodology employs a qualitative exploratory multi-case study: the method used has consisted in crossing the information currently known on the most advanced communication technologies (ICT) with the requirements of enhancing museum services, in order to determine the possible trajectories of applying the former to the latter.

The implementation of recommender system outlines the main implications and effects of an advanced market-driven digital orientation, as the system’s users are the starting point for innovation and the creation of value. For a museum, it will be possible to access to an additional system of knowledge alongside that of its scientific staff. This process has profound implications in the way in which a museum presents itself and how it is perceived by its visitors and, in a wider sense, by the potential demand.

The paper consists in an exploratory effort to introduce an analytical framework for an evolved adaptive museum orientation system; the empirical investigation can be structured in the inductive-predictive view of assessing this promising debate further.

Implementing the IoT blueprint entails introducing a plethora of new products, services and business models, opening new routes to guide and direct cultural events. Now, more than ever, sustainable development involves an intrinsic balancing act between the pluralism of data and that of customer needs, which is achieved through the elaboration of digital data.