Co-Training by Committee: A Generalized Framework for Semi-Supervised Learning with Committees

Many data mining applications have a large amount of data but labeling data is often di cult, expensive, or time consuming, as it requires human experts for annotation.Semi-supervised learning addresses this problem by using unlabeled data together with labeled data to improve the performance. Co-Training is a popular semi-supervised learning algorithm that has the assumptions that each example is represented by two or more redundantly su cient sets of features (views) and additionally these views are independent given the class. However, these assumptions are not satis ed in many real-world application domains. In this paper, a framework called Co-Training by Committee (CoBC) is proposed, in which an ensemble of diverse classi ers is used for semi-supervised learning that requires neither redundant and independent views nor di erent base learning algorithms. The framework is a general single-view semi-supervised learner that can be applied on any ensemble learner to build diverse committees. Experimental results of CoBC using Bagging, AdaBoost and the Random Subspace Method (RSM) as ensemble learners demonstrate that error diversity among classi ers leads to an e ective Co-Training style algorithm that maintains the diversity of the underlying ensemble.     

Acoustically generated flows in microchannel flexural plate wave sensors: Effects of compressibility

Acoustically generated flowfields in flexural plate wave sensors filled with a Newtonian liquid (water) are considered. A computational model based on compressible flow is developed for the sensor with a moving wall for pumping and mixing applications in microchannels. For the compressible flow formulation, an isothermal equation of state for water is employed. The velocity and pressure profiles for different parameters including flexural wall frequency, channel height, amplitude of the wave and wave length are investigated for four microchannel height/length geometries. It is found that the flowfield becomes pseudo-steady after sufficient number of flexural cycles. Both instantaneous and time averaged results show that an evanescent wave is generated in the microchannel. The predicted flows generated by the FPWs are compared with results available in the literature. The proposed device can be exploited to integrate micropumps with complex microfluidic chips improving the portability of micro-total-analysis systems.     

Stability and Exponential Synchronization of High-Order Hopfield Neural Networks with Mixed Delays

This paper investigates the problems of stability and synchronization for high-order recurrent neural networks with mixed delays. Firstly, we establish sufficient conditions to ensure the asymptotic stability and then the exponential synchronization. Furthermore, our results are applied to two chosen systems to demonstrate the effectiveness of the obtained theoretical results.     

Miscible blends of styrene–cinnamic acid copolymers with poly(ethyl methacrylate)

Differential scanning calorimetry and inverse gas chromatography have been used to investigate the miscibility behaviour of blends of poly(ethyl methacrylate) (PEMA) with styrene–cinnamic acid statistical copolymers PSCA5, PSCA8, PSCA23 having compositions of, respectively, 5, 8 and 23 mol% of cinnamic acid. Several probes with different chemical nature and polarity have been used to determine the polymer–solute and polymer–polymer interaction parameters. DSC and CPGI measurements indicate that poly(ethyl methacrylate) is miscible with each poly[(styrene)‐co‐(cinnamic acid)] copolymer as established from the observation of a single composition‐dependent glass transition temperature. This deduction is corroborated by the IGC data; comparison of the experimental retention volume of the blend with the algebraic average retention volumes of the pure components, together with negative values of the apparent polymer–polymer interaction parameter, establish the miscibility of the studied systems. Furthermore, the polymer–polymer interaction parameters are found to show marked probe dependence; this is discussed in terms of the Δχ effect. As indicated by the variation of the glass transition temperature with blend composition, the application of the Kwei and the Schneider approaches to the calorimetric results suggests the occurrence of strong specific interactions within the blends; the strength of these intermolecular interactions increases with the cinnamic acid content in the PSCA copolymer. © 2001 Society of Chemical Industry     

Open shop scheduling problem with a multi-skills resource constraint: a genetic algorithm and an ant colony optimisation approach

The continuous evolution of manufacturing environments leads to a more efficient production process that controls an increasing number of parameters. Production resources usually represent an important constraint in a manufacturing activity, specially talking about the management of human resources and their skills. In order to study the impact of this subject, this paper considers an open shop scheduling problem based on a mechanical production workshop to minimise the total flow time including a multi-skill resource constraint. Then, we count with a number of workers that have a versatility to carry out different tasks, and according to their assignment a schedule is generated. In that way, we have formulated the problem as a linear as and a non-linear mathematical model which applies the classic scheduling constraints, adding some different resources constraints related to personnel staff competences and their availability to execute one task. In addition, we introduce a genetic algorithm and an ant colony optimisation (ACO) method to solve large size problems. Finally, the best method (ACO) has been used to solve a real industrial case that is presented at the end.     

Reliability optimization of a redundant system with failure dependencies

In a multi-component system, the failure of one component can reduce the system reliability in two aspects: loss of the reliability contribution of this failed component, and the reconfiguration of the system, e.g., the redistribution of the system loading. The system reconfiguration can be triggered by the component failures as well as by adding redundancies. Hence, dependency is essential for the design of a multi-component system.In this paper, we study the design of a redundant system with the consideration of a specific kind of failure dependency, i.e., the redundant dependency. The dependence function is introduced to quantify the redundant dependency. With the dependence function, the redundant dependencies are further classified as independence, weak, linear, and strong dependencies. In addition, this classification is useful in that it facilitates the optimization resolution of the system design. Finally, an example is presented to illustrate the concept of redundant dependency and its application in system design. This paper thus conveys the significance of failure dependencies in the reliability optimization of systems.     

Intra- and Interorganizational Knowledge Services to Promote Informed Sustainability Practices

To promote sustainable construction, the construction industry needs to intensify its efforts to move to a knowledge intensive mode. Sustainability goals can only be achieved if new resources of knowledge and expertise inform construction activities. Some of this comes in the form of good practice and standards, but much will have to come from situated and contextual appreciations of sustainability goals and local practices developed across organizational, professional, and multicultural boundaries. The paper describes a knowledge management environment based on the web-services model aimed at promoting informed sustainability practices. This validation of the knowledge environment provided the basis for formulating recommendations for promoting knowledge-based sustainable construction.     

Berries anthocyanins as potential SARS-CoV–2 inhibitors targeting the viral attachment and replication; molecular docking simulation

The viral respiratory disease, severe acute respiratory syndrome (SARS), has turned into a global health concern. Till now, there is no drug or vaccine has yet been specifically approved for SARS-CoV-2. One of the urgent solutions against the recent COVID-19 disease is the use of dietary molecules, which can be found abundantly in functional food. In the current study, we have conducted a molecular docking approach for eighteen dietary molecules belong to the subclass of anthocyanins, as potential inhibitors of the main protease and spike glycoprotein of SARS-CoV-2. Both selected targets, playing a vital role in attachment and replication of the virus. The results indicated that cyanidin-3-arabinoside exhibited the lowest binding energy and located onto the pocket through a sufficient number of hydrogen bonds with the main protease virus. However, pelargonidin-3-glucoside and pelargonidin 3-rhamnoside display significant binding energy with the spike glycoprotein of SARS-CoV-2. All compounds mentioned above shown high drug-likeness and fulfils the Lipinski’s rule of five, as well as confer favorable toxicity parameters, in addition to ADME values. Considering the obtained results, regular consumption of berry fruits, which are rich in anthocyanin compounds, should be supportive to inhibit viral infectious by reducing of propagation and pathogenicity of SARS-CoV–2.     

NOx formation in post-flame gases from syngas/air combustion at atmospheric pressure

Species concentration measurements specifically those associated with nitrogen oxides (NO_x) can act as important validation targets for developing kinetic models to predict NO_x emissions under syngas combustion accurately. In the present study, premixed combustion of syngas/air mixtures, with equivalence ratio (Φ) from 0.5 to 1.0 and H₂/CO ratio from 0.25 to 1.0 was conducted in a McKenna burner operating at atmospheric pressure. Temperature and NO_x concentrations were measured in the post-combustion zone. For a given H₂/CO ratio, increasing the equivalence ratio from lean to stoichiometric resulted in an increase in NO and decrease in NO₂ concentration near the flame. Increasing the H₂/CO ratio led to a decrease in the temperature as well as the NO concentration near the flame. Based on the axial profiles above the burner, NO concentration increases right above the flame while NO₂ concentration decreases through NO₂-NO conversion reactions according to the path flux analysis. In addition, the present experiments were operated in the laminar region where multidimensional transport effects play significant roles. In order to account for the radial and axial diffusive and convective coupling to chemical kinetics in laminar flow, a multidimensional model was developed to simulate the post-combustion species and temperature distribution. The measurements were compared against both multidimensional computational fluid dynamics (CFD) simulations and one-dimensional burner-stabilized flame simulations. The multidimensional model predictions resulted in a better agreement with the measurements, clearly highlighting the effect of multidimensional transport.     

Particle swarm optimization algorithm for a vehicle routing problem with heterogeneous fleet, mixed backhauls, and time windows

Today, companies need to collect and to deliver goods from and to their depots and their customers. This problem is described as a Vehicle Routing Problem with Mixed Linehaul and Backhaul customers (VRPMB). The goods delivered from the depot to the customers can be alternated with the goods picked up. Other variants of VRP added to VRPMB are Heterogeneous fleet and Time Windows. This paper studies a complex VRP called HVRPMBTW which concerns a logistic/transport society, a problem rarely studied in literature. In this paper, we propose a Particle Swarm Optimization (PSO) with a local search. This approach has shown its effectiveness on several combinatorial problems. The adaptation of this approach to the problem studied is explained and tested on the benchmarks. The results are compared with our previous methods and they show that in several cases PSO improves the results.